OBSERVED CLASSROOM BEHAVIORS AND PERSONALITY TYPES OF 178 BEGENNING TEACHERS :. Them for the Degree of Ph D ~ g} - MICHIGAN STATE UNIVERSiTY ' a} 'SALLY ANN BRICKNER, o s. P 1970 wet: ‘ A m .m: .. LI-B RY Mkmngmw University v “Inf t-x—v- THEEIQ. This is to certifg that the thesis entitled M BEHAVIORS AND OBSERVED CLASSROO NG TEACHERS i PERSONALITY TYPES OF 178 BEGINNI presented 1313 Sally Ann Brickner, O.S.F. has been accepted towards fulfillment of the requirements for Ph. D. degree in Education ,7 eL/ V Major professor Date July 22, 19“) 0—169 \ llllllu’i ABSTRACT OBSERVED CLASSROOM BEHAVIORS AND PERSONALITY TYPES OF 178 BEGINNING TEACHERS BY Sally Ann Brickner, O.S.F. This study was designed to investigate the rela— tionship between the organizational structure of self— perceived attitudes, needs, and beliefs of beginning teachers and direct observations of overt behaviors in their elementary school classrooms. It was conjectured that accurate reports of self—perceptions and unbiased records of overt classroom behaviors would be congruent. That is, personalities of teachers would be consistent With the learning environments observed in their class— rooms. Answers to the following questions were sought: 1. What personality types.can be identified by in— vestigating the patterns of response of beginning teachers to items regarding their attitudes toward children and teaching, their needs and manner of believing? Sally Ann Brickner 2. What learning environments are observable in the elementary school classrooms of these teachers? 3. To what extent are the personalities of teachers consistent or congruent with overt behaviors ob— served in their elementary school classrooms? 4. Do the learning environments observed in elemen— tary school classrooms of beginning teachers differ between male and female teachers or the separate grade levels they teach? Data were gathered in two phases from 178 parti— cipants in the Elementary Intern Program of Michigan State University. Subjects entering.EIP in Fall, 1967 and Winter. 1968 responded to three personality inventories: the Minnesota Teacher Attitude Inventory. the Edwards Per- §9nal Preference Schedule, and a short form of the Rokeach Dogmatism Scale. Approximately l—l/2 years later, thirty- Seven trained observers recorded overt classroom behaviors Of these individuals during the final six weeks (May—June. 1969) of their teaching internship in public elementary Schools of southern Michigan. The Observation Schedule Sally Ann Brickner and Record (OSCAR III) was employed as a standardized re- cording form for each of the six observations per intern teacher. A random sample of eighty-eight subjects was drawn from the population. Q-type factor analysis was employed to identify persons who had similar score patterns on the ~MWAI. fourteen manifest needs of the EPPS (Absement was omitted from the analysis) and the Rokeacthoqmatism §ggle. Two personality types were identified by analyzing the factor loadings of individuals on the first rotated solution which accounted for 92% of the total variance. The characteristics of teachers assigned to each person— ality type were determined by means of weighted standard scores. Regression coefficients for each of the sixteen personality variables were computed by means of unre- stricted least squares. On the basis of the regression coefficients the personality type of each intern in the sample was predicted. One hundred forty—six interns were classified as Self-Centered; thirty—two interns were identified as Well—Integrated. The one hundred categories of overt classroom be— haviors were also submitted to factor analysis (R-type) 3 Sally Ann Brickner using principal components and orthogonal rotations. The fourth rotated solution accounting for twenty—five percent of the total variance was selected for further study. Using the factor loadings, five learning environments were described: Supportive. Conventional, Independent, Enriched, and Hostile. Factor scores for each of the 178 intern teachers on each of the five learning environments were computed. Relationships between teacher-personality types and learning environments identified in the factor analyses were hypothesized. Multivariate analysis of variance was used to test significance of differences in mean vectors of each personality type across the five learning environ— ments. Sex and teaching level differences were also in— vestigated. A fiducial level of five percent for rejec—_ tion of the null hypothesis was selected. Scheffe' post hoc comparisons of mean discriminant scores completed the investigation of.significant findings for sex and teaching levels. The following conclusions were derived from the investigation: 1. Sally Ann Brickner Factor analytic techniques could be employed to parsimoniously describe personality and classroom observation variables. Learning environments of Well-Integrated intern teachers did not differ significantly from learn- ing environments of Self—Centered intern teachers identified in this study. Learning environments of male intern teachers of this sample differed significantly from those of female intern teachers. Learning environments in classrooms of male teachers were observed as more Supportive, Conventional, and Enriched while those of female intern teachers were observed as more Independent and Hostile. Learning environments were significantly different among various teaching levels for this.sample of elementary intern teachers. Grades 6-8, 3-5, and 3-8 were observed to have more Enriched, Suppor— tive, and Conventional learning.environments than K—2 classes. Grades K—2 and Special Education 5 Sally Ann Brickner classes combined were observed as more Enriched, Hostile, and Independent learning environments than grades 6-8 or grades 3w8. The latter teaCh— ing levels showed more Supportive learning en— vironments. OBSERVED CLASSROOM BEHAVIORS AND PERSONALITY TYPES OF 178 BEGINNING TEACHERS BY Sally Ann‘Brickner, O.S.F. A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY College of Education 1970 g- 954 942 lie-aw” ”7/” Copyright by SALLY ANN BRICKNER, O . S .F . 1971 ACKNOWLEDGEMENTS The encouragement and support of many individuals facilitated my advanced study. Dr. W. Robert Houston provided his counsel and guidance in the writing of this thesis. He gave his time generously, expressed his con— fidence frequently, and allowed me to explore alternate approaches in solving the problem. No words can express the gratitude which is his due. I deeply appreciate the assistance of Dr. George Myers, Dr. Lawrence Sarbaugh, and Dr. Dale Alam who served on my doctoral committee. I was inspired most of all by the warmth and understanding of these gentlemen._ The Sisters of St. Francis of the Holy Cross merit runlimited thanks for their continuous support through prayer and letter, telephone calls, and personal visits. Their dedication inspires me to witness that same love and simplicity in my future life with them. To so many others, Elementary Intern Consultants, friends at Michigan State, and family, I am grateful. Because they listened, assisted, or encouraged me this thesis was completed. I.IIIII-III-IIII-------—————_—————-——’_"““‘-——__—__—‘fi“W’“*__'_—__"I ””” ‘“”m“““? ‘=_ .,_I_.:,' TABLE OF CONTENTS Page ACKNOWLEDGEMENTS. . . . . . . . . . . . . . . . . . iii LIST OF TABLES. . . . . . . . . . . . . . . . . viii LIST OF FIGURES . . . . . . . . . . . . . . . . . . xii LIST OF APPENDICES. . . . . . . . . . . . . . Xiv Chapter I. NATURE OF THE INVESTIGATION . . . l l Theoretical Basis for the Study . . . 1 Need for the Study. . . . . . . . . . . . 4 Scope and Definition of the Problem . . 7 Basic Assumptions . . . . . . . . . . . . 8 Summary of Procedures . . . . . . . . . . 10 The Study Sample. . . . . . . . . . . . .10 Collection of Data. . . .'. . . . . . . 10 Analysis of Data. . . . . . . . . . 12 Definition of Terms . . . . . . . . . . . ' 15 Organization of the Report. . . . . . . . 18 II. REVIEW OF RELATED RESEARCH. . . . . . . . . 20 Introduction. . . . . . . . . . . . . . . 20 Personality Theory. . . . . . . . . . . . 21 iv It FIlIIIIIfIIIIIII----—-_—————__———___'S=g§“_——__————”7W"“I—____—l‘” TABLE OF CONTENTS.—-Cont. Chapter Attitudes . . . . . . . . . . . . . Needs . . . . . . . . . . . . . Beliefs . . . . . . . . . . . . . . ‘ Summary . . . . . . . ..... . . . . ObServation of Overt Classroom Behaviors. Summary . . . . . . . . . . . . . . . III. PROCEDURES UTILIZED IN THE INVESTIGATION. Introduction. . . . . . . . Selection of the Study Population . . Description of the Study Sample . . . Teaching Assignments. . . Age Characteristics . . . . . . . . Educational Background. . . . . . . Community Background. . . . . ‘. Instrumentation . . . . . . . . . . The Minnesota.Teacher Attitude __ Inventory . . . . . . . The Edwards Personal Preference Schedule. . . . . . . . . . . The Dogmatism Scale . . . . . . . The Observation Schedule and Record Procedures of the Investigation . 0 Collection of Personal and Personality Data. . . . . . . . . . . . . Collection of Data on Overt Classroom Behaviors . . . . . . . . . . . . Training the Observers. . . . . Preparation of Data for Analysis. . Analysis of Data. . . . . . . . .. 69. 69 69 7O 79 84 86 89 93 TABLE OF CONTENTS.——Cont. Chapter Factor Analysis . . . . . . . . R—type Factor Analysis. . . . . . . . Q—type Factor Analysis. . . . . . . . Multivariate Analysis of Variance . . Summary . . . . . . . . . . . . .. IV. RESULTS OF THE FACTOR ANALYSIS. . . . . . Introduction. . . . . . . . . . . . Personality Types . . . . . . . . . . . Type I: The Self—Centered Personality. Type II; The Well—Integrated Personality . . . . . . . . . . . Classroom Learning Environments . Dimension I: Supportive Learning Environment . . . . . . . . . . . Dimension II: Conventional Learning Environment . . . . .. . . . . . . Dimension III: Independent Learning Environment . . . . . . . . . . Dimension IV: Enriched Learning Environment . . . . . . . . . . . Dimension V: Hostile Learning Environment . . . . . . . . . Summary . . . . . . . . . . . . . . . . V. RESULTS OF MULTIVARIATE ANALYSIS OF VARIANCE. . . . . . . . . . . . . . Introduction. . . . . . . . . . . . . . Personality Types and Sex Related to Five Learning Environments. . . . . . vi Page 105 109 114 117 120 122 122 122 129 132 134 136 139 144 144 145 TABLE OF CONTENTS.--Cont. Chapter Page Teaching Level Related to Five Learning Environments. . . . . . . . . . . . . . 153 Summary . . . . . . . . . . . . . . . . . 165 VI. CONCLUSION OF THE INVESTIGATION 167 Introduction. . . . . . . . . . . . . 167 Summary of the Investigation. . . . . . 167 Limitations of the Investigation. . . . 172 Data Collection . . . . . . . . ... . 172 Direct Observation of Overt Classroom Behaviors . . . . . . . . . . . . . . 173 Statistical Techniques. . . . . . . . . 175 Conclusions . . . . . . . . . . . . . .176 Implications of the Investigation . . . . 178 Teacher Eddcation . . . . . . . . 180 Recruitment and Placement . . . . . . . 182 Teacher Supervision and In—Service Education . . . . . . . . . . 183 Suggestions for Further Research. . 184 ‘ SELECTED BIBLIOGRAPHY . . . . . . . 188 APPENDICES. . . . . . . . . . . . 202 vii 4—. Table 3‘.\]: 3.2 3.3 3.4 3.6 3.7 3.8 3.9 LIST OF TABLES Page Sex and Teaching Assignment of 178 Intern Teachers in the Elementary Intern Program, 1968—1969 . . . . . . . . . . . . 72 Distribution of 178 Intern Teachers by Age . 73 Age at the Time of Orientation Toward and Commitment to Teaching for the 178 Intern Teachers. . . . . . . ‘ 74 Type of School Attended by Intern Teachers during Their Elementary and High School Years. . . . . . . . . . . . . . . . . ._ 75 College Background of 178 Intern Teachers Prior to Admission to the Elementary Intern Program . . . . . . , , 76 Grade Point Average Reported by 178 Intern Teachers at Entry to the Elementary Intern Program, 1967 . . . . . . . , 77 Type of Community Background Reported by 178 Intern Teachers, 1968—1969 . . . . , 78 Intercorrelations Among Three Factor Scales Based on OSCAR Scores of Forty—Nine Beginning Teachers . . . 92 Stages in Analysis of Personality Variables. 101 Stages in Analysis of 6h; Hundred Overt Classroom Behavior Variables . . 103 viii LIST OF TABLES.-—Cont. Table 3.11 4.1 4.2 4.4 4.7 4.8 4.9 5.1 Page Percent of Variance Explained by Each Factor in the Eight Rotated Solutions of 100 Classroom Observation Variables. Subjects with Highest Loading on Factor d. Subjects with Highest Loading on Factor B. Descending Array of Standard Scores for Each of Sixteen Variables in the First Personality Type (Factor d). . . . . Descending Array of Standard Scores for Each of Sixteen Personality Variables in the Second Personality Type (Factor B) Descending Array of Differences Between Standard Scores for Each of Sixteen Personality Variables in Factors d and B . Variables Loading Highest on Factor I (Supportive Learning Environment). . . . , Variables Loading Highest on Factor II (Conventional Learning Environment). . Variables Loading Highest on Factor III (Independent Learning Environment) . . . Variables Loading Highest on Factor IV (Enriched Learning Environment). . . . Variables Loading Highest on Factor V (Hostile Learning Environment) subclass Frequencies of Male and Female Elementary Intern Teachers in Two Personality Types. . . ix 111 125 127 128 131 133 138 139 140 141 142 . 145 LIST OF TABLES.--Cont. Table 5.2 Mean Products: Multivariate Analysis of Variance of Learning Environments for Personality Types by Sex . . . . . . . 5.3 Tests of Significance in Multivariate Analysis of Variance for Two Personality Types of Male and Female Elementary * Intern Teachers. . . . . . . . 5.4 Subclass Means of Male and Female Elemen- tary Intern Teachers in Two Personality Types. . . . . . . . . . . . . . . . . . . 5.5 Step—down Test of Sex Effect . . . . . . . . 5.6 Subclass Frequencies of Male and Female Elementary Intern Teachers at Four Teaching Levels. . . . . . . . . . . 5.7 Mean Products: Multivariate Analysis of Variance of Learning Environments for Sex by Teaching Level. . . . . . . . . . . . 5.8 Tests of Significance in Multivariate Analysis of Variance for Male and Female Elementary Intern Teachers at Four Levels of Teaching. . . . . . . . . . . . . . . . 5.9 'SubClass'Means cf Male and Female Elementary Intern Teachers at Four Teaching Levels. 5.10 Mean Discriminant Scores for Four Teaching Levels . . . . . . . . . . . . . . . . 5.11 Step-down Test of Teaching Level Effect. 5.12 Post Hoc Comparisons of Means for Teaching Levels Using the First Discriminant Score. X Page 146 147 149 152 153 155 156 158 161 162 164 IJST OF TABLES.--Cont. Table 5.13 Post-Hoc Comparisons of Means for Teaching Levels Using the Second Discriminant Score. . . . . . . . . . . . . . . . Appendix A.1 Means and Standard Deviations of OSCAR ‘ Categories: Pupil Activities. . . . A.2 Means and Standard Deviations of OSCAR Categories: Teacher Activities. . A.2 Means and Standard Deviations of OScAR (Categories: Grouping Activities . . A.4 Means and Standard Deviations of OScAR Variables: Materials Used by the Teacher and Pupils . . . . . . . . . A.5 Means and Standard Deviations of OScAR Variables: Subject Matter Observed. A.6 Means and Standard Deviations: Personality variables. . . . . . . . . . . . A.7 Factor Loadings for One Hundred Classroom Observation variables on Each of Five Factors in the Fourth Rotated Solution xi Page 165 230 231 232 233 233 234 236 LIST OF FIGURES Figure Page 2.1 Interrelationships Between Attitudes, Needs, and Beliefs Within the Person— ality, and the Press of Internal and External Forces on the Personality . . . . 60 3.1' Generalized Factor Matrix. . . . . . . . . . 108 3.2 Score Matrix of One Hundred Classroom Observation variables for 178 Subjects . . 110 3.3 Correlation Matrix of One Hundred Classroom Observation variables for 178 Intern Teachers . . . . . . . . . . . . . . . . . 110 3.4 Scree Slope Indicating the Minimum Number of Factors to be Extracted from the Factor Matrix Based on 100 Overt Classr room Behaviors . . . . . . . . . . . . . . 113 3.5 Transposed Matrix of Personality Scores for 178 Elementary Intern Teachers . . . . . . 115 3.6 Intercorrelation Matrix of Eighty-Eight Intern Teachers. . . . . . . . . . . . . . 115 3.7 Scree Slope Indicating the Minimum Number of Factors to be Extracted from the Factor Matrix Based on the Personality Data . . . . . . . . . . . . . . . . . . . 116 5.1 Comparison of Overall Means for Male and Female Elementary Intern Teachers on Each of the Learning Environments. . . . . 150 xii LIST OF FIGURES . —-Cont . Figure Page 5.2 Comparison of Overall Means for Teaching Levels of Elementary Intern Teachers on Each of the Dependent Variables. . . . . . 159 xiii LIST OF APPENDICES Appendix Page A. THE MANIFEST NEEDS ASSOCIATED WITH EACH OF THE FIFTEEN EDWARDS PERSONAL PREFERENCE SCHEDULE VARIABLES . . . . . . 203 B. ROKEACH DOGMATISM SCALE (SHORT FORM). . . . 208 C. OBSERVATION SCHEDULE AND RECORD (OSCAR III) MANUAL AND RECORDING FORM. . . . . . 212 D. MEANS AND STANDARD DEVIATIONS FOR VARIABLES ON THE OBSERVATION SCHEDULE AND RECORD AND PERSONALITY VARIABLES. . . 229 E. FACTOR LOADINGS FOR ONE HUNDRED CLASSROOM OBSERVATION VARIABLES ON EACH OF FIVE FACTORS IN THE FOURTH ROTATED SOLUTION. . 235 xiv CHAPTER I NATURE OF THE INVESTIGATION Theoretical Basis for the Study Every teacher is an impact teacher.l What he knows is important; so is what he does. But his whole self, his personality has the real impact.2 The learning environment of children is touched, for good or for ill, by ”who and what" the teacher is. How do we know what a teacher is like? Snygg and Combs suggest that the outsider must observe the indi— vidual and infer from that observed behavior the phenom— enal self.3 Admittedly, such observation merely approxi- mates the real phenomenal self which only the individual * lRaymond H. Reno, The Impact Teacher (St. Paul: 3M Education Press, 1967) 53. 2J. W. Getzels and P. W. Jackson, "The Teacher‘s Personality and Characteristics," Handbook of Research on Teaching, ed. by N. L. Gage (Chicago: Rand McNally, 1963). 506. 3All perceptions of the self a person has at a particular instant is called the phenomenal or perceived self. knows.4 However, observations of overt behaviors can be supplemented with the individual's own perceptions of himself, thus providing as complete a picture of a person as limited humans can obtain. Therefore two forms of observation of behavior can be distinguished: indirect and direct. The former in— cludes personal reflections, reactions, expressions by an individual regarding his own attitudes, interests, beliefs, needs, motives, and actions. The latter concerns the per- ceptions by an outside source of the overt behavior of the individual.5 To the extent that beliefs, attitudes, interests, and needs of an individual are congruent with-his overt behavior we can say that the person behaves consistently. If an individual accurately reports his perceptions of himself and if an outsider makes unbiased observations of overt behaviors of the same person, the indirect and di— rect observations should be congruent or consistent. L 4Donald Snygg and Arthur W. Combs, Individual Be— havior (rev. ed.; New York: Harper & Row, Publishers, Inc., 1959), 44. " > 5David G. Ryans, Characteristics of Teachers (Washington, D.C.: American Council on Education, 1960), 21—22. A. This theory of congruency is applicable to teachers in their work. If the teacher perceives himself as having positive attitudes toward children and teaching or sees himself as an Open—minded individual, we would expect a learning environment in his elementary school classroom to reflect that perception. On the other hand, the teacher who perceives himself as having negative views of children and teaching or is more dogmatic probably maintains a learning environment which portrays these attitudes. Researchers indiCate that teacher personality is related to classroom social interaction. Bowers and Soar declare: The analysis of classroom social interaction can proceed best if attention is directed to the personality characteristics of teacher and pupils. Personality traits, or as identifiable, personality patterns, are covariants in studying the differential effect on pupils of varying de- grees or qualities of classroom interaction. Personality traits condition, modulate, promote certain responses from pupils; they activate, direct, formulate pupil reactions in the class— room learning situation. They are basic to teachers and pupils working together success— fully in some quest for knowledge, skills, understanding and attitudes. It seems inade- quate to consider classroom social interaction independent of the teacher and pupil person— ality traits. The present study was designed to investigate the rela— tionship between self—perceived attitudes, needs, and beliefs of teachers and their overt teaching behaviors. Need for the Study Numerous teacher-personality studies have been reported in educational literature.7 After reviewing re— cent reports, Getzels and Jackson summarized their find- ings. DeSpite the critical importance'of“the prob— lem and a half—century of prodigious research effort, very little is known for certain about the nature and measurement of teacher person— ality, or about the relation between teacher personality and teaching effectiveness. The regrettable fact is that many of the studies so 6 , _ . .. Norman D. Bowers and Robert S. Soar, "The Influ- ence of Teacher Personality on Classroom Interaction," QQprnal of Experimental Education, XXX (June, 1962), 309. 7 . . A. S. Barr, "The Measurement and Prediction of Teaching Efficiency: A Summary of Investigations," figurnal of Experimental Education, XVI (June, 1948), , 203-283. S. J. Domas and D. V. Tiedman, "Teacher Compe— tence: .An Annotated Bibliography," Journal of Experi- Egntal Education, XIX (December, 1950), 99—218. far have not produced significant results. Many others have produced only pedestrian findings. For example, it is said after the usual inventory tabulation that good teachers are friendly, cheerful, sympathetic, and morally virtuous rather than cruel, depressed, unsympathetic, and morally depraved. But when this has been said, not very much that is especially useful has been revealed. For what conceivable human interaction-—and teaching implies first and foremost a human interac- tion—-is not the better if the people involved are friendly, cheerful, sympathetic, and vir— tuous rather than the opposite. Distinctive features of teacher personality and of the effective teacher remain unknown, Getzels and Jackson concluded. Nor does a dearth of recent classroom observation research exist though methodology for such studies may still be in its infancy.9 Numerous classroom observation systems have been developed and applied to research on teaching.10 8Getzels and Jackson, "The Teacher's Personality and Characteristics,” 574. 9Donald M. Medley and Harold E. Mitzel, "Measur- ing Classroom Behavior by Systematic Observation," Hand- :gbk of Research on Teaching, ed. by N. L. Gage (Chicago: Rand McNally, 1963), 297—325. 0Anita Simon and E. Gil Boyer, ed., Mirrors.for Behavior (Philadelphia: Research for Better Schools, Inc., Classroom Interaction Newsletter, January, 1968). Few studies combine teacher personality and sys— tematic observation of overt teaching behaviors. Existent reports usually focus on one dimension of teacher person- ality such as authoritarianism, dogmatism, attitudes, needs. But if personality means the teacher as a unique whole or refers to the dynamic organization of attitudes, needs, and beliefs of the teacher then the more dimensions included in a study, the more information gained about the teacher's personality. Furthermore, the structure of teacher personality more than any one dimension would seem to be related to overt teaching behaviors.ll If common organizational structures of teacher personality could be identified, if overt classroom be- haviors of teachers having similar personality profiles were congruent with their eXpressed attitudes, needs, and beliefs, then selection of candidates and prediction of their effectiveness in teaching would be facilitated. The constantly increasing number of applicants for teacher k llIra J. Gordon, Relationships Between Personality Variables and Classroom Behavior of Teaching Interns (Gainsville, Florida: University of Florida, 1964), U.S Department of Health, Education, and Welfare, Project Number 1717, 8—9, education programs, the shortage of funds for educating teachers, the saturation of teachers on the job market require an answer to the selection problem. Scope and Definition of the Problem This study investigated the relationship between the organizational structure of self-perceived attitudes, needs, and beliefs of beginning teachers and direct ob- servations of overt behaviors in their elementary school classrooms. Answers to the following questions were sought: ‘7 1. What personality types can be identified by in— vestigating the patterns of response of beginning ‘ teachers to items regarding their attitudes toward children and teaching, their needs and manner of believing (indirect observations)? What learning environments are observable in the elementary school classrooms of these teachers (direct observations)? '1 : 3. To what extent are the personalities of teachers consistent or congruent with overt behaviors ob— served in their elementary school classrooms? Sex of the teacher and teaching level: primary, inter- mediate, upper-elementary, or special education, could be relevant variables. Besides studying congruency be- tween indirect and direct observations of teacher be- haviors, another question was investigated: 4. Do the learning environments observed in elemen- tary school classrooms of beginning teachers differ between males and females or among the separate grade levels they teach? Basic Assumptions Several basic assumptions must be made when theo- rizing about the relationship between self—perceptions of teachers and direct observations of their overt teaching behaviors. Teachers can accurately report perceptions of their attitudes and interests, needs, and manner Of believing. Paper—pencil inventories can be used to validly assess the attitudes, needs, and manner of be— lieving of beginning teachers. Teacher behaviors in their elementary school classrooms are observable. Teacher behavior is characterized by some degree of consistency. Teacher behaviors are classifiable qualitatively and quantitatively. Relatively unbiased observations of overt be— haviors of teachers in their elementary school classrooms can be made if an objective form for recording behaviors and adequately trained ob— servers are employed. 10 Summary of Procedures The Study Sample The sample for this study included 178 pre—service teachers who participated in the Elementary Intern Program (EIP) of Michigan State University during 1967—1969. These individuals elected EIP after completing approxi— mately two years of college. The last two years of their teacher preparation program consisted of: (l) a pre— intern year devoted to academic work in liberal arts, study of methods of teaching and student teaching, and (2) a year of internship during which each person assumed full re3ponsibility for instruction Of‘pUpilS in a public school affiliated with MSUeEIP. During this final year, an intern consultant guided five or siX intern teachers. Collection of Data _________________ During the first week of their methods courses and Prior to contact with children in elementary school 11 classrooms, the 178 pre-interns (EIP enrollees for 1967— 1968) completed the following inventories: 1. Teacher Education Inventory—-an instrument de- vised to collect specific demographic data on students enrolled in the Elementary Intern Pro— gram. 2. Minnesota Teacher Attitude Inventory-(MTAI)——a personality measure designed to predict the type of social climate the teacher will maintain in the classroom. 3. Doqmatism Scale—7a short form of the instrument developed by Milton Rokeach to measure the open- closedness of cognitive belief systems. 4. Edwards Personal Preference Schedule (EPPS)—-an ipsative scale which purports to measure fifteen manifest needs of the reSpondent, and his consis_ tency in responding to the items. 12Descriptions of the personality inventories and the classroom observation form are included in Chapter III. 12 Thirty—seven intern consultants were trained in the use of the Observation Schedule and Record (OScAR III)13 by means of filmed episodes of classroom behaviors. The intern consultants observed and recorded their intern teachers' overt classroom teaching behaviors in six sep- arate observation periods during May and early June, 1969. One thousand sixty—eight observations (6 x 178) on one hundred variables were recorded. Both the personality and the classroom observation data were coded and keypunched on computer cards in pre— paration for analysis on the CDC 3600 computer at Michigan State University. Apalvsis of Data Two major questions of this study focused on (1) types of teacher personality and (2) kinds of learning environments found in elementary school classrooms of _‘ "Measuring Classroom Behavior l3Medley and Mitzel, A.sample'of,the by Systematic Observation," 278—286.“ OScAR III recording form and a description of categories is included in Appendix C. 13 these teachers. Factor analysis was employed to investi- gate these questions. To explore personality types among the intern teachers, Q—type factor analysis was used. Due to limi- tations in the size of the intercorrelation matrix which could be used in the computer program, eighty—eight intern teachers were randomly selected from the total sample. Fourteen EPPS scores, the NEAI score, and the dogmatism score for each of the eighty-eight subjects were factor analyzed to determine which individuals had similar re- sponse patterns on the paper—pencil personality measures. The first rotation yielded two personality types which accounted for ninety—two percent of the total variance. The characteristics of teachers assigned to each person— ality type were determined by means of weighted standard- ized scores. Regression coefficients for each of the sixteen variables included in the factor analysis were computed by means of unrestricted least squares. On the basis of the regression coefficients the personality type of each intern in the sample was predicted. One hundred forty-six interns were classified in the first personality type; thirty—two interns were grouped in the second type. 14 The one hundred categories of overt classroom behaviors from the Observation Schedule and Record (OScAR III) were also submitted to factor analysis (R—type) using principal components and orthogonal rotations. The fourth rotated solution was selected for further study. Using the factor loadings for each factor in this rotation, five learning environments were described: Supportive, Conven— tional, Independent, Enriched, and Hostile. Factor scores for each of the 178 intern teachers on each of the five learning environments were computed. Factor analysis thus provided a parsimonious de- scription of the data: 1. Two distinct teacher-personality types were iden- tified by analyzing the reSponse patterns of eighty—eight pre-interns on sixteen personality variables. 2. Five learning environments were described using the fourth rotated factor solution of overt classroom behaviors of the intern teachers. Relationships between teacher—personality types and learning environments identified in the factor 15 analyses were hypothesized. Multivariate analysis of var— iance was employed to test significance of differences in mean vectors of each personality type across the five classroom learning environments. Sex and teaching level differences were also investigated. A fiducial level of five percent for rejection of the null hypothesis was selected. Post hoc comparisons of discriminant scores completed the investigation of the data. Definition of Terms Those terms related to the problem in general will be explicated first of all. Then the particular variables associated with self—perceptions of the interns and direct observations of their overt teaching behaviors will be defined. Finally, an explanation Of Specific terms re— lated to the Elementary Intern Program of Michigan State University will be presented. Self:perception.-—Personal reflection, reaction, or expression of an individual regarding his attitudes, interests, beliefs, needs, motives, actions. V"— 16 Direct observation.——Perception by an outside source of the overt behavior of an individual. Teacher behavior.-—The behavior, or activities of persons as they go about doing whatever is required of teachers, particularly those activities which are con— cerned with the guidance or direction of the learning of others. Attitude.——Disposition of mind with regard to a fact or state; a feeling or emotion toward a fact or state. Personality.——Totality of an individual's behav— ioral and emotional tendencies unique to himself; the organization of the individual's distinguishing character traits, attitudes, needs, or habits. Doqmatism.——Degree of openness or closedness in a Person's systems of belief and disbelief.1 N§§§,——Condition of want, desire for fulfillment When some lack prevails. Internship (in teacher education).—-A professional preparation program for teacher candidates in which x l4Milton Rokeach, The Open and Closed Mind (New York: Basic Books, Inc., 1960), 6. “*figgev dzasr’ V‘ in l7 interns are contracted by local school boards, assigned a carefully planned teaching load for a school year, and supervised by a highly competent teacher released from teaching to devote full time to the supervision of intern teachers.15 Elementary Intern Program (EIP).——A cooperative endeavor of personnel in public school districts, com— munity colleges, and Michigan State University which pro— vides academic, professional, and practical preparation Of elementary teachers in a four—year program designed to integrate theory and practice in teaching. EIP Centers.-—Ten geographic locations in southern Michigan (Alpena, Battle Creek, Bay City—Saginaw, Detroit, Grand Rapids, Lansing, Livonia, Macomb, Pontiac, and Port Huron) each with a staff consisting of a center director and intern consultants who maintain relations with the cooperating school districts. Qgpperatinq School Districts.——Fifty-one public school districts associated with various EIP centers and \ 15Internships in Teacher Education (Washington, D-C-= The Association for Student Teaching, 47th Year— book, 1969), xi.. 18 contracting intern teachers and consultants for the school year. Intern.—-Full-time teacher having complete reSpon— sibility for learning in a particular subject area and/or at a specific grade level, salaried by the school board of the c00perating school district, and supervised by an intern consultant. Intern Consultant.——Clinical supervisor selected for competency in teaching and supervisory work salaried by the school board of the cooperating school district to aid five or six intern teachers. Center Director.——Michigan State University faculty member and coordinator of the EIP program at one of ten locations in Michigan. Organization of the Reporp This study will be reported in the five succeed— ing Chapters. Literature pertaining to attitudes, needs, I and beliefsiof teachers and direct observations of overt teaching behaviors will be reviewed first. Salient char— acteristics of the study sample, instrumentation. and 19 procedures will be presented in Chapter III. Analysis of data will be discussed in two chapters: the results of factor analyses of personality and classroom observation data (Chapter IV); and the findings of multivariate analysis of variance (Chapter V). The final chapter of the thesis summarizes the investigation, enumerates con— clusions, and discusses limitations and educational im— plications of the study. CHAPTER II REVIEW OF RELATED RESEARCH Introduction Two major tOpics are discussed in this chapter. Under personality theory, the attitudes, needs, and beliefs of teachers are reviewed. Since congruency between felt attitudes, needs, and beliefs of teachers and their overt behaviors in the classroom is the focus of this research, effort was made to report research studies incorporating direct observations of overt teaching behaviors with assessment of personality fac- tors. In the second section of this chapter, the method of direct observation of overt classroom be— haviors is discussed. The problems associated with classroom observation research are studied and relevant research is reviewed. 20 21 PersonalityATheory Theorists view personality in two ways. Their vantage may be the individual in his life space or the individual among many individuals. The first probes in— dividual differences, is ideographic. The second seeks general principles, is nomothetic.l Actually, the views are complementary rather than contradictory. The focus of this research is the individual in the group, the teacher among his colleagues. The atti- tudes, needs, and beliefs of a teacher integrate within his personality Structure. He is unique. But his inte- grated personality tends to be more like some teachers‘ and less like others. In short, there are typical ele— ments in the personality structures of many individuals. Topologies, then, may be found in a group of teachers. Though the elements of personality are interre— lated, each factor investigated in this study will be considered individually. Attitudes, needs, and manner of believing will be discussed and representative re— search presented. k lRobert-M. Allen, Variables in Personality Theory 2nd Personality Testinq(Springfi§Id, Illinois: Charles (L Thomas Publisher, 1965), 4-10. 22 Attitudes Like personality, the concept of attitude has varied definitions. In this study, attitude will mean a "tendency or disposition to evaluate an object or the symbol of that object in a certain way."2 In other words, an attitude indicates what a person's habitual mode of behavior will be. Valuation of objects ranges on a favorable-unfavorable continuum; that is, attitudes are termed favorable—unfavorable. A teacher possesses attitudes toward any thing or person connected with his profession. If his attitudes are positive or favorable, the teacher should be able to maintain a state of harmonious relations with his pupils charac— terized by mutual affection and sympathetic understanding. The pupils should like the teacher and enjoy school work. The teacher should like the children and enjoy teaching.3 2 . . D. Katz and E. Stotland, "A Preliminary Statement to a Theory of Attitude Structure and Change," in S. Koch, ed., Psychology: A Study of a Science, Vol. III, 1959, 428. Quoted in Chester A. Insko, Theories of Attitude Change (New York: Meredith Publishing Co., 1967), 2. 3Walter W. Cook, Carroll H. Leeds, and Robert Callis, "Predicting Teacher—Pupil Relations," The Evalua— Eion of Student Teaching (Washington, D.C.: The Associa— tion for Student Teaching 28th Yearbook, 1949), 67. 23 On the other hand, a teacher with generally unfavorable attitudes toward children may dominate the classroom en— vironment. If he is successful, a tense or fearful atmo— sphere may prevail in a situation which appears orderly. If he is unsuccessful, his nervousness and frustration may result in chaotic classroom conditions. The Minnesota Teacher Attitude Inventory was de— veloped to assess those attitudes of a teacher which predict how well he will relate interpersonally with children and how satisfied he will be with the teaching profession.4 Since its inception, many researchers have utilized the MTAI either in studies of the measuring de- vice itself or as a predictive instrument. Description of the inventory and discussion of reliability, validity, and other pertinent characteristics of the test can be found in Chapter III. Attention will now be directed toward reported research on attitudes of the teacher as related to classroom behaviors. Medley and Mitzel probed factors which might account for differences in (1) effectiveness of beginning 4Walter W. Cook, Carroll H. Leeds, and Robert Callis, Minnesota Teacher Attitude Inventory Manual (New York: The Psychological Corporation, 1951), 2. , 24 _teachers, and (2) persistence in teaching as a career among teacher education graduates. The MTAI was one of several tests administered to 343 education students. Forty—nine beginning teachers (of whom only three were male) were selected for further study on the effective— ness of their teaching. These individuals taught in nineteen public elementary schools in New York City. The Observation Schedule and Record (OScAR 2a) was used by six observers to record overt classroom behaviors. Twelve observations were made for each teacher. Attempts to relate observation results with predictor data gathered when the teachers were in college met with little success. The three factor scales (Emotional Climate, Verbal Em— phasis, and Social Structure) seemed to account for some of the differences in teacher—pupil rapport and princi- pals' ratings, but not for differences in pupils' achieve— ment.5 A questionnaire was sent to the 343 individuals three years after their student teaching experience. Those who persisted in teaching were found to be L 5Medley and Mitzel, "Measuring Classroom Behavior by Systematic Observation," 278—282. 25 relatively older, prepared for elementary rather than secondary teaching, and "satisfied" with student teaching as they had experienced it. A significant decrease in yaverage MEAI score occurred over the three-year period. Two factors seemed to be associated with the decline: a shift from more extreme to less extreme Opinions, and an increase in emphasis on the importance of limits to pupil behavior and maintenance of high academic standards.6 The Minnesota Teacher Attitude_Inventory was one of several predictive measures used by Seibel when study- ing one hundred graduate students enrolled in a one—year Masters in Education Program. Seibel questioned the pos— sibility of predicting the classroom behaviors of these individuals during their student teaching experience. The subjects were observed by the regular classroom teachers and their college supervisors. Following the observations, ratings of eight Specific classroom be— haviors were made on a Likert-type scale on which values could vary from 1 (does not exhibit) to 7 (seizes every __ 6Donald M. Medley, Harold E. Mitzel, and'W. Rabinowitz, "Longitudinal Studies of a Group of Teacher Education Graduates," Journal of Teacher Education, X (March, 1959), 117-119. 26 opportunity to exhibit). A cannonical correlation of .59 was obtained between the predictor and criterion vari— ables. Seibel indicated that teacher behaviors charac- terized by seriousness, compliance with pupil requests, high emotional support and affectionate physical contact, allowing pupils to do things for themselves, soliciting suggestions, and some immobility combined to provide max— imum predictability. Giving rewards contributed little. A teacher exhibiting these behaviors tended to have many previous leadership activities with children, a high score on the Minnesota Teacher Attitude Inventogy, and a low score on the Paranoia Scale. He also viewed many pupil misbehaviors as serious. Evidence of a substantial rela— tionship between antecedent variables and the classroom behaviors of student teachers existed, Seibel concluded.7 Ragsdale explored the relationship of change in student teachers' attitudes toward children‘s behavior and teacher—pupil relations with change in student teachers' classroom behavior during a ten—week period of E 7 . . . D. W. Seibel, "Predicting the Classroom Behavior of Teachers," Journal of Experimental Education, XXXVI (Fall, 1967), 26—32. 27 student teaching. The MTAI, California F~Scale, and Ryans' Teacher Characteristics Schedule were administered to forty-nine elementary education student teachers at Anderson College, Indiana, before and after student teaching. Two trained observers used Flanders” Interac— ’ tion Analysis and Ryans’ Teacher Characteristics Classroom Record (Patterns X0, Y0, and 20) to record eight observa- tions of the student teachers. No significant changein student teachers' attitudes concerning children and teach— ing as expressed on the predictive measures was found. A 1 significant positive change occurred in the student teachers' classroom behaviors as measured on the Class— room Observation Record; no significant change was indi— cated by the Interaction Analysis data. Student teachers Spent less time talking, and learned to ask questions eliciting creative responses and ideas from the pupils, observed Ragsdale. These findings indicate that classroom behavior may change significantly without a corresponding change in expressed attitudes when the same aspects of the teaching process are being measured. ‘_ 8Elva Mae Ragsdale, "Attitude Changes of Elemen- tary Student Teachers and the Changes in Their Classroom Behavior During Student Teaching9(unpublished Ed.D. dis- sertation, Ball State University, 1967), Dissertation m, XXVIII (August, 1967), 521-522 A. 28 Bridgman sought to identify characteristics which distinguish teachers as a group and which differentiate between effective and ineffective teaching among the teachers of New Hanover County, North Carolina. Personal and professional characteristics, attitudes, values, and creativeness were assessed. Classroom performance was measured using Ryans' Classroom Observation Record on selected teachers during the 1964e1965 school year. Product moment correlations, twtests, and analysis of variance were used to analyze the data. Findings relevant to attitudes and classroom performance were: 1. Elementary teachers had more positive attitudes than their colleagues; females had more positive attitudes than males. 2. Teachers with positive attitudes were more crea— tive, permissive, and social, gave higher grades, were more active professionally, and were rated higher on performance. 3. Teachers with high performance ratings had higher positive attitudes, were less 29 authoritarian, more creative, and had lower eco— nomic values. To discover relationships between attitudes and certain teaching behaviors, Friedman studied a sample of twelve beginning teachers, the twelve c00perating teachers with whom they served their internship, eight laboratory school teachers with whom they worked, and the two univer- sity teachers by whom they were instructed. The MTAI was used to assess teacher attitudes. Four two—hour observa- tions were recorded by a trained observer on a specially devised Observation Checklist to describe teacher behavior. The university teachers and participating laboratory school teachers were found to have more positive attitudes toward children than beginning or c00perating teachers. The observed behaviors of the former groups were less controlling, and encouraged greater independence and as— sumption of reSponsibility by the learners than did * 9John Northan Bridgman, "Selected Teacher Charac— teristics and Their Relationships with Certain Behavior Patterns and Teaching Effectiveness" (unpublished Ed.D. dissertation, The University of North Carolina at Chapel Hill, 1967), Dissertation Abstracts, XXVIII (March, 1968), 3524 A. 3O behaviors of the other two groups. Significant differ- ences between behaviors of beginning and participating teachers, between beginning and university teachers were found when overall results were considered. No such sig— nificant differences occurred between beginning and co- Operating teachers.lO Furr studied the effects of feedback from obser— vations on the verbal and nonverbal behavior of thirty— six female student teachers in the elementary school. A parallel part of his study explored the relationship be— tween scores on the MTAI administered before and after student teaching. No significant difference was found between the experimental and control groups as a result of feedback from observations of (1) indirect and direct verbal behavior, or (2) encouraging and inhibiting nonm verbal behavior. A positive correlation was found between direct verbal and inhibiting nonverbal classroom behavior. A negative shift in eXpressed attitudes toward children 10L. R. Friedman, "An Investigation of Certain Teaching Behaviors of and Influences Upon New Elementary Teachers? (unpublished Ed.D. dissertation, The University of Florida, 1967), 149 pages, Dissertation Abstracts, XXIX (July, 1968), 165 A. 31 and teaching occurred in both groups, with a significantly greater negative shift shown by the experimental group. Furr concluded that instruction in the nature and use of category systems for recording classroom behaviors prob- ably was not sufficient in itself to bring about signifi— cant change in verbal and nonverbal classroom behavior of elementary school student teachers. Research was conducted by Bowers and Soar with fifty-four volunteers from two elementary school syttems in Tennessee. The eXperimenters examined differential effects of laboratory and human relations training on teachers. Four attitude and personality inventories were completed by the subjects preceding and following labora— tory and human relations training of teachers in the ex- perimental group. Overt classroom behaviors of these teachers were recorded on Medley and Mitzel‘s Revised 2 Observation Schedule and Record.1 llOneta Roberts Furr, "Effects of Observational Feedback on Verbal and Nonverbal Classroom Behavior Of Student Teachers" (unpublished Ed.D. dissertation. North Texas State University, 1968), Dissertation.Abstracts, XXIX (April, 1969), 3495 A. -‘ 2Norman D. Bowers and Robert S. Soar, Studies of flhman Relations in the TeachipgflLearning Process: V: Einal Report: Training for Classroom Teachers (Nashville, Tennessee: Vanderbilt University, 1961), 210 pages. , _._...___.____b_( is. 32 In an early phase of the study, Rippy investigated the relationships between overt classroom behaviors and teacher attitude and personality characteristics. Anal— ysis of variance around the regression line and Pearson product moment correlations were used to test differences. Six departures from linearity and nineteen correlations significantly different from zero were observed. His study demonstrated the value of observing specific be— haviors in the classroom. The psychopathic deviate, schizophrenia, and hysteria, psychastnia and hypochondri- asis scales of the Minnesota Multiphasic Personality In: ventory, the MTAI, and Bowers Teacher Opinion Inventory were significantly related to identifiable aSpects of classroom behaviors. The first four scales had the most predictive power, however. The way in which the teacher described herself was reflected more generally in measures of pupil behavior than in measures of both teacher and pupil classroom behavior.l k 13Mark Leo Rippy, Jr., "Certain Relationships Be- tween Classroom Behavior and Personality Characteristics of Selected Elementary Teachers" (unpublished Ed.D. dis- sertation, George Peabody College for Teachers, 1960), 240 pages. \ , 5 ) I ! I 1 33 A coherent picture of personality resources basic to skillful teacher—pupil interpersonal relationships was outlined by Bowers and Soar on the basis of their human relations study: . . Skillful interaction with pupils requires on the part of the teacher reSponsibility, and depth of affective relationship; it requires that she be well enough adjusted that much of her energy is not drained off in dealing with her own intrapersonal tensions; and she must be able to perceive herself and others clearly and represent herself honestly in communication with others. A teacher must, in short, care; must not have this concern blocked by her own intrapersonal tensions; and must be relatively free of distorting mechanisms, and able to enter honestly into relations with others. Perhaps what this reduces to is that a teacher must be able to use her "self" openly, clearly and honestly in her interactions with pupils.14 How do we identify such an individual prior to entry into the profession? This question was analyzed in much of the research previously cited. Wilk and his col- leagues were also concerned with this problem. In a care— fully designed study, they explored the relationship be- tween a student teacher's needs for integration and dOmi- nance, attitudes, and other personality variables and E 4Bowers and Soar, "The Influence of Teacher Per- sonality on Classroom Interaction,“ Journal of Experimen- Sal Education, xxx (June, 1962), 311. 34 ratings of their classroom behaviors made by supervisors and trained observers. The Miller Analogies Test, the Cooperative Reading Test, the MMPI, MTAI, and Bowers Teacher Opinion Inventory comprised the psychological measures administered prior to admission. Counselors' judgments were also used as a basis for admission to the teacher education program. 7 Thirty-eight female student teachers, selected from 167 individuals who participated in the interviews, were randomly assigned to public elementary schools in four geographic regions in a midwestern city. Each of five trained observers visited each student teacher once a semester. Two semesters of student teaching were re— quired, one in lower and one in upper elementary teaching. Ten Observation reports for each student teacher were collected during the two semesters of the study. Two methods of systematic observation employed were the 9b: servation Schedule and Record (OScAR III) and the Minne_ §Qta System of Interaction Analysis. The OScAR items were grouped into eight scales: Teacher—Class Activities, Teacher-Individual Activities, Pupil Disruptive Behavior, Pupil—Class Activities, Pupil-Individual Activities, 35 Teacher—Grouping Activities, and Teacher and Pupil Use of Materials. The researchers hypothesized that a teacher's need for integrating behavior was related to indirect verbalizations and her need for dominance was related to direct verbalizations. Interaction analysis was used to assess indirect and direct teacher talk. The major find— ings related to the current investigation were: 1. Teachers were more often observed giving positive attention and encouragement to individual pupils during their upper ele- mentary teaching than when in lower ele— mentary teaching. 2. At the upper level, teachers more often used direct questions or answered a pupil's question rather than ignoring it. 3. As student teachers gained experience, they took more active roles in directing the class, and students became more pas— sive participants. 4. The variety of activities in which pupils led the class Was greater when student teachers taught in the lower grades rather than upper grades. 5. Student teachers and pupils made more use of a variety of materials when teaching in lower grades rather than upper grades. 6. Student teachers who preferred upper grade levels had significantly better Classroom Emotional Climate scores when they taught in the upper grades. Those who preferred the lower grades showed no change in scores at the different levels. 36 7. Student teachers preferring the lower grades had a higher mean Classroom verbal Emphasis score than those preferring upper grades. All student teachers, re— gardless of preference, had higher Verbal Emphasis scores when they were in the upper grades. 8. Significantly less disruptive pupil be- havior occurred in classrooms where the level and variety of pupil-class activ- ities were high, and where the student teachers' work with the individuals was ObServed to be interested, helpful and supportive. 9. Student teachers' need for integration and security could be predicted from MTAI scores and sophomore honor point ratio. Students who had greater command of the subject matter and who had certain atti— tudinal dispositions allowed students to respond freely. They could accept pupil feeling, could support, praise, and en— courage pupils. This research again supported the supposition that person- ality factors relate to the behaviors of elementary teachers in their classroom. 15Roger E. Wilk, William H. Edson, Don DaVies, and Naomi Chase, A Study of_§he Relationship Between Ob_ §§rved Classroom Behaviors of Elementary Student Teachers, Eradictors of These Behaviors, and Ratings by Supervisors (Minneapolis, Minnesota: University of Minnesota, 1962), 78 pages plus appendices. 37 Eases Murray heightened interest in the construct of 16 needs in his major work, EXplorations in Personality. He defined a need as: . . . a construct which stands for a force in the brain region, which organizes . . . action so as to transform in a certain di— rection an existing, unsatisfying situation Several principles basic to Murray's personality theory are incorporated in this definition. The brain unifies man's personality since "need" centers in the brain re— gion. This "need," whether a force or an emotional charge, moves a man in a defined direction to reduce his dissatisfaction, whether internally or externally created. When man generates his own tension, he eXpresses his need for positive thrust, excitement, or movement. By reducing his tension, man expresses his need for homeostasis. Murray established a taxonomy of needs that moti— vate man. For example, abasement or achievement, play or sex, exhibition or autonomy might be the forces moving man —‘ l6Henry'A. Murray, Explorations in Personality (New York: Oxford, 1938). 17Ibid., 123. 38 to behave in a particular manner. Needs were classified into five types: primary and secondary, proactive and re— active, overt and covert, focal and diffuse, effect and modal. Primary and Secondary needs: This distinction refers to states of develop— ment rather*than degree of importance. Pri- mary needs develop prior to the secondary needs. Primary needs are of a biological or organic nature while secondary needs are psy- chogenic. Achievement and affiliation are examples. Proactive and reactivegpeeds: Proactive need originates within the person, while reactive need originates in man's en— vironment. In the latter condition, man reacts to a stimulus. Proactive-reactive needs are bound together in a cause—effect relationship. Qvert_andyCoyert Needs: This distinction refers to needs which can be openly expressed (overt) as opposed to those which must be hidden (covert). Society may sanction achievement but censure aggres— sion, for instance. Focal and diffusegneeds: A focal need centers on one particular object or person for satisfaction. Diffuse need, such as man's need to avoid pain, covers a wide range of objects. 18Ibid., 152—226. 39 Effect and modal needs: A need which has a direct or specific goal is termed an effect need. But the process by which we achieve the effect may equal or ex- ceed the importance of achieving that effect. This is termed a modal need. In this study, the secondary or psychogenic needs of beginning teachers are examined. As a background, major studies which identify specific needs of teachers will be reviewed first. Then several reports on the re- lationships between needs of beginning teachers and overt classroom behaviors will be summarized. Many of the studies reported utilize the Edwards Personal Preference Schedule.20 The fifteen needs which this instrument pur— ports to measure relate directly to Murray's need theory. Jackson and Guba examined the manifest needs of 366 public school teachers in a Midwestern city. The sample of teachers included 118 males (91 secondary and 27 elementary) and 218 females (52 secondary and 196 1 . . . ‘ 9Ledford J. Bischof, Interpreting_Personality Theories (New York: Harper and Row, Publishers, 1964), 140—177. 20A. L. Edwards, Edward Personal Preference Sched— 21§= Manual (Rev. ed.° New York: The Psychological Cor— poration, 1959). A description of the fifteen needs mea— sured on this inventory is contained in Appendix A. 4O elementary). Edwards Personal Preference Schedule was administered to the subjects and the scores compared to the college norm groups on whom the test was standardized. Teachers showed significantly greater need for deference and lower need for heterosexuality than did the liberal arts norm group. Male secondary teachers and female teachers eXpressed significantly greater need for order and endurance, and significantly less need for exhibi— . 2 . tion. 1 "These five needs,” concluded the researchers, "appear to be more or less typical of teachers in general, at least insofar as the teacher Occupational group may be different from liberal arts students."22 The same sample of teachers was divided into groups on the basis of teaching eXperience (0—3 years, 4—9 years, and 10 or more years) and sex. The needs of novice, intermediate, and veteran teachers were rank~ ordered for each subject. Male and female veteran teachers had markedly similar profiles; the rank order correlation was .70. Deference and order ranked highest, _4.‘ 21Philip W. Jackson and Egon G. Guba, "The Need Structure of In—Service Teachers: An Occupational Anal— ysis." Sghool Review, va (June, 1957), 176—192. 22Ibid., 178. 41 and heterosexuality the lowest for this level of teaching experience. Male and female novice teachers had quite dissimilar need structures (rho = .19). .Young female teachers resembled older female teachers (rho = .71) more closely than young male teachers' need patterns resembled . 23 their older male counterparts (rho = .32). Sheldon and others hypothesized that individuals scoring high on a number of ”warm teacher scales" ought to have need structures different from teachers scOring very low on several of the same scales. The EPPS was one of several tests administered to twenty students (ten having very high scores and ten very low scores on the "warm teacher" tests). Among the Six needs which were examined, ”potentially good teachers" had significantly higher scores on affiliation and dominance and signifi— cantly lower scores on aggression, succorance, and abase— 24 ment. Thus the hypothesis of their study was confirmed. Need structures of student teachers who preferred lower elementary school classes were compared to the need _— 2 3Ibid., 184. 2 4M. S. Sheldon, J. M. Coale, and R- COpple, "Cur— rent Validity of the 'Warm Teacher Scales,'" Journal of Educational Psychology, L (February, 1959), 37—40. 42 structures of persons preferring upper elementary classes. Southworth found that abasement, affiliation, succorance, and nurturance were expressed as higher manifest needs of students who preferred the lower grades. Achievement, aggression, and exhibition were greater needs of students preferring later elementary classes. Anderson was interested in comparing the need pro— files of teachers in schools identified as having open organizational climates compared to schools with closed organizational climates. Eight elementary schools in a large southern school district were used in the study. School climate was determined by administering Halpin and Croft's Organizational Climate Description ggestionnaire to the faculty members. Seventy—one teachers taught in schools having open organizational climates, and fifty- five teachers taught in schools having closed climates. No significant differences in need profiles were observed. Analysis of extreme cases did not reverse the finding. When each need was examined by means of t—tests, teachers 25Horton C. Southworth, "A Study of Certain Per- sonality and value Differences in Teacher Education Majors Preferring Early and Later Elementary Teaching Levels" (unpublished Ed.D. dissertation, Michigan State Univer- sity, 1962). 43 in open organizational climates were found to have signif— icantly less need for intraception and abasement than faculty members in closed organizational climates. The manifest needs of student teachers were com- pared to their rapport with pupils during student teach- ing. Ninety—one student teachers completed the EPPS prior to their student teaching eXperience. The Pupil Reaction Inventory, a questionnaire designed to indicate teacher- pupil rapport, was administered to their pupils following student teaching. Medley found no significant relation— ship between manifest needs of student teachers and pupil- teacher rapport. He then studied the twenty-five indi— viduals who scored less than eleven points on the consis- tency scale of the EPPS. He hypothesized that these teachers were being honest and expressing their true needs, while the other sixty—six student teachers knew how to fake on personality tests. A correlation of .76 between the needs of "inconsistent" respondents and _— 6Donald Anderson, "A Comparison of Edwards Per— sonal Preference Schedule Patterns of Elementary School Teachers in Open and Closed Organizational Climates" (un- published Ed.D. dissertation, Auburn University, 1966), 104 pages. AlSo see, Donald D. Anderson, "Personality Attributes of Teachers in Organizational Climates,” Jour- Qal of Educational Research, LXII (July, 1969), 4417443. 44 pupil—teacher rapport was found. Intraception, achieve— ment, aggression, and abasement correlated positively with teacher—pupil rapport and heterosexuality correlated nega- tively. Beta coefficients for the first four needs were Significant at the .01 level and for heterosexuality at the L05 level. Here, concluded Medley, is a picture of the successful teacher as she perceives herself.27 Using sociometric nominations of beginning educa— tion students at North Texas State University, Clary sep- arated ninety student teachers into three groups of indi- viduals exhibiting aggressive, submissive, or normal be— havior patterns. Scores on the Edwards were then obtained for these ninety student teachers. Criterion profiles of need patterns for each group were developed on the basis of mean scores on the fifteen scales of EPPS. Finally, the twenty student teachers whose needs profiles most closely corresponded to the criterion profiles were rated by their college supervisor and supervising teacher as aggressive, submissive, or normal. Individuals identified as aggressive had a significantly greater need for ¥ 27Donald M. Medley, "Teacher Personality and Teacher Pupil Rapport,” Journal of Teacher Education, XII (June, 1961), 152—156. 45 deference than did submissive or normal student teachers. Subjects in this study did not differ significantly from the norm group of the EPPS. No significant relationship between needs profiles of student teachers and ratings of . . 28 their classroom behaV1ors was observed. Four dimensions of teacher behavior were studied by'Travers and his associates: 1. Achievement: degree of emphasis on aca— demic activities and the importance of excellence; 2. Affiliation: degree of interaction with students and emphasis on warm personal relationships; 3. Control: degree to which the moment—to- moment behavior of students is controlled by the teacher; 4. ReCOgnition: degree to which the teacher is the center of attention.29 These dimensions of behavior were assessed by two methods: a sampling of one hundred verbal statements of the teacher, and observer ratings of thirty—five variables on a 28Eldon Gandy Clary, "Predicting Student Teaching Behavior From Needs Profiles by Comparison with Sociomet- rically Defined Groups" (unpublished Ed.D. dissertation, North Texas State University, 1968), 94 pages, Disserta— tion Abstracts, XXIX (April, 1969), 3488—3489 A. 9Norman E. Wallen, Robert M. W. Travers, Ian E. Reid, and Kenneth H. Wodtke, "Relationships Between Teacher Needs and Teacher Behavior in the Classroom," Qernal of Educational Psychology, LIV (February, 1963), 23. 46 seven—point scale. Two samples of elementary school teachers (seventy—seven from an urban setting and forty- one from a rural setting) responded to a questionnaire and were observed in their classrooms. The questionnaire was scored on the four dimensions already mentioned, and the results compared to data from the observed behaviors. Control and Affiliation were positively related on all measures of teacher behavior. A significant negative correlation was found between the same needs as viewed by the observers. The Control need as measured on the questionnaire correlated positively with the controlling behavior of the teacher. Such behavior was related to lack of warmth and, to a lesser degree, lack of confidence on the part of the teacher. The authors concluded that the best predictive measure of performance is some type of instrument in which the teacher can answer straight and simple questions about his typical behaviors.3O 0Robert M. W. Travers, Norman E. Wallen, Ian E. Reid, Kenneth H. Wodtke, Measured Needs of Teachers and Their Behavior in the Classroom (Salt Lake City: Univer-. sity of Utah, 1961, Final Report to U.S. Department of Health, Education and Welfare, Office of Education Con— tract No. 444 [8029]). 47 Gordon hypothesized that the need structure of a teacher is directly related to overt teaching behaviors; that the need structure is more important than any Single need. She studied a sample of fifty elementary, twenty— three junior high, and twenty-nine senior high student teachers of the University of Florida. Randomization was not employed in the selection process. The Edwards Per— sonal Preference Schedule, Thurstone Temperament Schedule, and two case studies were administered to the subjects. The Emotional Climate Scale of OScAR 2a was used in re— cording classroom behaviors. Three subscales included manifest teacher hostility, supportive teacher behavior, and disorderly pupil behavior. Means and standard devia— tions of the scores were computed, all variables were intercorrelated, and then submitted to varimax rotations. Three common factors (Leadership, Nurturant Affiliation, and Emotional Climate) were obtained for males, and secondary and elementary females. Dominance had a high loading for males and for secondary females; females in general had impulsive tendencies. Male subjects eXhibited less need to relate to people in an emotionally supportive and helpful way. Elementary teachers engaged in more supportive ingly, ele1 tility tha that facto variable a ality seem ferential teaching 1, servation Could be a eVer. Emot When compa lnSvlll epartmem: 48 supportive behaviors than secondary teachers. Surpris- ingly, elementary female teachers manifested more hos— tility than female secondary teachers. Gordon concluded that factor analytic procedures rather than a variable—by— variable approach to the measurement of teacher person— ality seems to offer some support to the concept of dif— ferential personality organization. However, sex and teaching level should be analyzed separately.31 The ob— servation schedule (OScAR) proved useful since observers could be adequately trained without undue effort.32 How— ever, Emotional Climate seemed to be too broad a construct when comparing overt behaviors to personality character- istics; subscales indicated differences more clearly. Beliefs Just as Murray theorized that an individual pos— sesses a system of needs, Rokeach theorized that a person has a unique system of beliefs. 31 . . Ira J. Gordon, Relationships Between Personality Variables and Classroom Behavior of Teaching Interns (Gainsville, Florida: University of Florida, 1964), U.S, Department of Health, Education, and Welfare, Project Number 1717, 127. 32Ibid., 131. The E the be, consci< given ‘ lives a seril single sets, , that t1 given ‘ For exampl, the best f ing monarc‘ 0f governm Sp SYStem. o disbelief nor belief gious, eth ilar belie ilmy. T it is not YOU believ \ 33 York: 3&5 34 49 The belief system is conceived to represent all the beliefs, sets, expectancies, or hypotheses, conscious and unconscious, that a person at a given time accepts as true of the world he lives in. The disbelief system is composed of a series of subsystems rather than merely a single one, and contains all the disbeliefs, sets, expectancies, conscious and unconscious, that to one degree or another, a person at a given time rejects as false. For example, positively affirming belief in democracy as the best form of government for our country implies negat— ing monarchy, oligarchy, and all other types as our form of government. Specific to the individual is his belief_disbelief system. On the other hand, the open-closedness of belief— disbelief systems is restricted to neither individuals nor beliefs: political, economical, philosophical, reli— gious, ethnic, or scientific. Individuals having dissim— ilar belief—disbelief systems sometimes act or react sim— ilarly. This phenomenon led Rokeach to say that ”. it is not so much what you believe that counts, but how you believe."34 33Milton Rokeach, The Open and Closed Mind (New York: Basic Books, 1960), 33. 34Ibid., 6. uates and with the governed 1 less by i: with a mo: distingui t0 the si- . 3 nation. 5 R sYstem as (a) a 0f be organ about Provi ance 4 Three Set of dogmat tarlanism \ matiSm’ ll tiSm: PS Ch010 (Number 1 50 A person having a more open belief system eval— uates and acts on information on its own merits, in accord with the structural requirements of the situation. He is governed more by his inner self—actualizing forces and less by irrational inner forces. Conversely, a person with a more closed belief system has greater difficulty distinguishing between substantive information (pertaining to the situation) and nonsubstantive or irrelevant infor— mation. Rokeach designates the relatively closed belief system as more dogmatic. He defines dogmatism as: (a) a relatively closed cognitive organization of beliefs and disbeliefs about reality, (b) organized around a central set of beliefs about absolute authority which, in turn, (c) provide a frameWOrk for patterns of intoler— ance.and qualified tolerance toward others.36 Three sets of variables are subsumed under the construct of dogmatism: closed cognitive systems, general authori— . . . 37 tarianism, and general intolerance. 35Ibid., 54-70. 36Milton Rokeach, ”The Nature and Meaning of DQg_ matism," Psychological Review, LXI (May, 1954), 194. 37 . . . . . Milton Rokeach, ”Political and Religious Dogma_ tism: An Alternative to the Authoritarian Personality," Psychological Monographs: General and Applied, LXX (Number 18, Whole Number 425, 1956), 4. ”an ____4 teaching? having cl haviors o the same would ten tion; dec fected; i fected by R thinking aChiEVeme cipal's r1 reported, 3 in? in Re 111% I' J- Lem tudes and (Septembe: 51 Can this cognitive belief theory be applied to teaching? It would appear that behaviors of teachers having closed systems of belief would differ from be— haviors of teachers having open systems of belief, given the same classroom situation. A more dogmatic teacher would tend to confuse relevant and irrelevant informa— tion; decisions about subsequent action would be af— fected; interpersonal relations would probably be af— fected by inappropriate action or reaction. Relationships between dogmatism and critical thinking skills, social class, values,and academic achievement, pre—decisional information search, prin— cipal's ratings of teachers, and other factors have been 3 . . reported. 8 ConcluSions from these studies are useful 38C. Gratton Kemp, "Improvement of Critical Think- ing in Relation to Open—Closed Belief Systems,” JOurnal of Experimental Education, XXXI (March, 1963), 321—323. I. J. Lehmann, ”Some Socio-Cultural Differences in Atti— tudes and values," Journal of Educational Sociology, XXXVI (September, 1962), 1—9. R. M. Frumkin, "Dogmatism, Social Class, Values, and Academic Achievement in Sociology," Journal of Educational SOCiology, XXXIV (May, 1961), 398— 403. Barbara H. Long, and Robert C. Ziller, ”Dogmatism and Predecisional Information Search," Journal of Applied Psychology, XLIX (October, 1965), 376—378. D. Musella, "Open—Closed—Mindedness as Related to the Ratings of Teachers by Elementary School Principals," Journal of Experimental Education, XXXV (Spring, 1967), 75—79. John T. Mouw, "Effect of Dogmatism on Levels of Cognitive Processes," Journal of Educational Psychology, LX (Octo— ber, 1969), 365—369. for gel ever, < will be school dogmat: summer to the open-m: sex, y( taught t0 dogi consid, With 0i 52 for generating hypotheses about teacher behaviors. How— ever, only reported research on the dogmatism of teachers will be reviewed here. Rabkin tested Soderbergh's assumption that public school teachers in America are excessively and unwittingly dogmatic.39 One hundred seven teachers registered for summer school at the University of Washington responded to the Rokeach Dogmatism Scale. The teachers were more open—minded than any of the norm groups of Rokeach. Age, sex, years of experience, religious affiliation, grades taught, and marital status were not significantly related to dogmatism. Teachers in this biased sample exhibited a considerably lower degree of rigid thinking when compared with other college and non—college groups.40 In a similar study, Cappelluzzo and Brine adminis— tered the Dogmatism Scale and a brief questionnaire to 254 undergraduates in education. The respondents (prospective teachers) were neither more nor less dogmatic than state 39P. A. Soderbergh, ”Dogmatism and the Public School Teacher," JOurnal of Teacher Education, XV (Sep— tember, 1964), 245—251. 40 . . n . Leslie Y. Rabkin, The Dogmatism of Teachers?,” Journal of Teacher Education, XVII (Spring, 1966), 47—49, M”— univer than t with v levels lated groups dOgmat subje( 51998: tribut effect Ceptua and p] XX (S) Their m 53 university students in general. They were more dogmatic than the in—service teachers measured by Rabkin. Students with various subject matter interests portrayed different . 41 levels of dogmatism. Ohnmacht administered Ryans' Teacher Character— istics Schedule, the California F—Scale, the Dogmatism Scale, and measures of analytic set to fifty-seven male secondary education majors. The scores were intercorre— lated and the matrix factor analyzed. Subjects were grouped on analytic set and dogmatism. Low analytic—high dogmatic individuals differed significantly from other subjects in their tendency to give information. Results suggested that open—mindedness and analytic set could con— tribute to an understanding of teacher behavior if their effects as moderating variables were investigated.42 Results of studies on authoritarianism and per— ceptual accuracy provide evidence that the perceiver's own character will affect the manner in which he perceives 41Emma M. Cappelluzzo and James Brine, "Dogmatism and Prospective Teachers," Journal of Teacher Education, XX (Summer, 1969), 148-152. 42F. W. Ohnmacht, "Teacher Characteristics and Their Relationship to SOme Cognitive Styles,” gpp£p§l_2£ Educational Research, LX (January, 1967), 201—204. others.‘ relatior supervis personal teachers complete (used t( percepti consists Score or on the I for that 0Pen—mir their p, ing teen Nor Were Curate j of Stude C<>“'Pare \ a O ,_. O 54 others.43 Brumbaugh and his colleagues investigated the relationship between dogmatism of student teachers and supervising teachers and their perceptions of the inter— personal needs of each other. Forty secondary school teachers and their public school supervising teachers completed the Dogmatism Scale and Schutz' FIRO—B Scale (used to measure interpersonal needs). The measure of perceptual accuracy in estimating interpersonal needs consisted of an index of difference between the total score on each dimension registered by the target person' on the FIRO—B protocol and the dimension score predicted for that person by the subject. Results indicated that open-minded student teachers were no more accurate in their perceptions of the interpersonal needs of supervis— ing teachers than were closed—minded student teachers. Nor were Open—minded supervising teachers any more ac— curate in their perceptions of the interpersonal needs of student teachers. Chi—square analysis was used to compare subject matter areas to level of dogmatism. Mathematics, science, and Social studies teachers were 43Paul F. Secord and Carl W. Backman, Social Psy— Shglggy (New York: McGraw—Hill Book Company, 1964), 80, found in an Openn their colleg highei Rating studer ilarit and t} rating “umber the di Studer, the MT Practi matiSm by See dict s \ H, Bei racy, u 3”~33 55 found to be significantly more closed—minded than teachers in areas of foreign language, English, or the fine arts.44 Johnson investigated the relationship between open—closed mindedness of student teachers and ratings of their student teaching made by cooperating teachers and college supervisors. Supervising teachers tended to give higher ratings to student teachers who were closed—minded. Ratings by college supervisors and the dogmatism scores of student teachers were not significantly related. The sim— ilarity of open— and closed—mindedness of student teachers and their cooperating teachers had little effect on the ratings of success in student teaching, but a significant number of student teachers changed in dogmatism score in the direction of their cooperating teacher. Low dogmatic student teachers expressed more favorable attitudes on the MTAI. Student teachers' own ratings of success in practice teaching were not related to their level of dog— matism. The degree of open—closed—mindedness as indicated by scores on the Dogmatism Scale could not be used to pre— dict success in student teaching if ratings of college ‘— 44R. B. Brumbaugh, Kenneth C. Hoedt, and William H. Beisel, Jr., "Teacher Dogmatism and Perceptual Accu— racy.“ Journal of Teacher Education, XVII (Fall, 1966), 332—335. supe crit sub; istj clos visc hig} era] supe feet that effe ratj C105 lish Tea< the] Tea( 56 supervisors and cooperating teachers were used as the criterion, Johnson concluded.45 These conclusions must be accepted with caution, though. Attrition among the subjects was considerable, and differential character— istics of the groups were not reported. Lewis examined the relationship between open— and closed-mindedness of teachers and perceived effectiveness of teaching (self—rating, principals' ratings, and super- visors' ratings). One hundred twelve junior and senior high school teachers participated in the study. In gen— eral, the level of dogmatism of a teacher was not related to self—rating of teacher effectiveness. Principals and supervisors tended to give higher ratings of teacher ef— fectiveness to more dogmatic teachers. Lewis suggested that dogmatism may have limited usefulness in identifying effective teachers.46 Perhaps not only dogmatism but ratings of effectiveness should be called into question. 45James Johnson, ”The Relationship of Open— and Closed—Mindedness to Success in Student Teaching" (unpub— lished Ed.D. dissertation, George Peabody College for Teachers, 1966), 125 pages. 46Franklin Garner Lewis, "The Relationship of Au— thoritarianism as Revealed by the Rokeach Dogmatism Scale and Perceived Effectiveness of Teaching as Indicated by Teachers' Self—Ratings, Principals' Ratings and dog tea and Uni tea thi her use the 57 Piele investigated the relationship of teacher dogmatism to verbalizations in the classroom under actual teaching conditions. Seventy teachers from elementary and junior high schools in the Northwestern part of the United States completed the Dogmatism Scale. Seventeen teachers from the upper and lower quartiles of the score distribution were selected for further study. These thirty—four teachers tape—recorded six twenty—minute seg— ments of regular class lessons. Observers trained in the use of Flanders' system of interaction analysis recorded the verbal behaviors. Open and closed—minded teachers appeared to differ significantly (a) in their monopoliza— tion of talk in the classroom and (b) in their use of a variety of verbal behaviors. Children in the classroom appeared to differ (a) in their use of student—to—student interaction, and (b) in their use of student—to—student interaction with silence or confusion following. Piele related his findings to Hanny's study of the relationship Supervisors' Ratings" (unpublished Ed.D. dissertation, North Texas State University, 1968), 90 pages, Disserta— tion Abstracts, XXIX (December, 1968), 1682 A. m 58 between level of dogmatism and teaching in a simulated situation. Closed—minded teachers appeared to use a wider variety of verbal behaviors and to monopolize talk in the classroom more under actual teaching conditions than did teachers in simulated conditions. Under simulated teach— ing conditions, open-minded teachers tended to use indi- rect influence more than did teachers in the normal class— room. Under both simulated and actual teaching condi— tions, students of open—minded teachers seemed to talk more than students of closed—minded teachers. Piele ad— vanced some plausible explanations for these results. Closed—minded teachers may be more concerned about class- room control, and try to discourage student talk by monopolizing classroom talk. In so doing, they use a variety of verbal behaviors, and obtain a higher score for indirect influence.47 47Philip Kern Piele, ”The Relationship of Teacher Open and Closed Mindedness to Classroom Verbal Behavior? (unpublished Ph.D. dissertation, University of Oregon, 1968), 88 pages, Dissertation Abstracts, XXIX (May, 1969), 3899-3880 A. k. Summa field conta ality tion t Attitl withix other, condit them, matica Struct “POD t Variab 8-9.— 59 Summary Despite the fact that many studies occurred in a field setting and that intervening variables could easily contaminate the work, results Show that teacher person— ality factors, when studied individually or in conjunc— tion with other factors, do relate to teaching behaviors. Attitudes, needs, and beliefs of a teacher integrate within his unique personality. They influence one an— other, and are affected by the press of environmental conditions. In a sense, it is artificial to separate them, as Gordon indicated.48 Figure 2.1 portrays sche— matically the dynamic relationships between these con— structs within a person and environmental forces acting upon the individual. 48Gordon, "Relationships Between Personality Variables and Classroom Behavior of Teaching Interns," 8—9.“' ' pi, 60 U = Environment Personality Abstract Nature Attitudes Individual Fig. 2.l.——Interrelationships Between Attitudes, Needs, and Beliefs Within the Personality, and the Press of Internal and External Forces on the Personality. see wha ex; tea beh vat lit and tie Lem PUp Sif Sys dim 61 Observation of Overt Classroom Behaviors Observation of overt behaviors in the classroom seems an obvious means of obtaining information regarding what does occur in teaching. However, use of systematic observations is beset with difficulties. Much time and expense can be expended to little avail. The privacy of teachers is invaded. The presence of an observer causes behaviors to be atypical. Past studies utilizing obser— vation of overt classroom behaviors have contributed little to our knowledge of "effective teaching." These and other objections are raised when systematic observa— tion of the teaching—learning process is advocated.49 DeSpite these objections, numerous category sys— tems for recording classroom behaviors of teachers and pupils have been developed.50 These systems can be clas— sified into three major categories: affectively oriented systems, cognitively oriented systems, and multin dimensional systems. In the affective systems, such as 49 . . ' Medley and Mitzel, ”Measuring Classroom Behavior by Systematic Observation," 247—248. 50 . . . . Anita Simon and E. Gil Boyer, ed., Mirrors for §§h§¥l2£ (Philadelphia: Research for Better Schools, Inc., Classroom Interaction Newsletter, January, 1968). Witha to te Cogn: beha‘ may 1 diffs ferex Class ence sEime tWee: If h, agreE A“ It her, 62 Withall's and Flanders', dimensions of teaching related to teacher—pupil interactive behaviors are considered. Cognitive systems of Taba, Smith, and Bellack attend to behaviors that shape habits of thinking, or methods of acquiring information and skills. Both affective and cognitive dimensions of behavior are considered in com— prehensive systems such as Medley and Mitzel's Observa— tion Schedule and Record. Selection of a particular system depends on the purpose for which the instrument is being employed.51 Systematic observations of classroom behaviors may be used to obtain reliable and valid measurements of differences in the typical behaviors which occur in dif- ferent classrooms, or in different situations in the same classroom. A measure is reliable if the average differ— ence between independent measurements obtained in the same classroom is smaller than the average difference be— tween two measurements obtained in different classrooms. If behaviors in a particular class are stable, observers agree on what does occur, items in the category system 51 B. Othanel Smith, "Recent Research on Teaching: An Interpretation," The High School Journal, LI (Novem— ber, 1967) , 63—65. 63 are consistent, and differences between different classes are large, reliability will be high. Valid measurement requires differences in scores to reflect differences in behaviors. Three conditions must be fulfilled; (l) a representative sample of the behaviors to be measured must be obServed; (2) an accurate record of the observed behaviors must be obtained; and (3) the records must be scored so as to faithfully reflect differences in be— havior.52 Reliability is a tricky concept. Though consis— tency of measurement is the commonly accepted meaning, many types of consistency can be measured. Behaviors of an individual may be the same (consistent) under repeated observations. Or, the individual may retain the same relative position in a group of individuals. Again, the percent of observer agreement or correlation between two sets of observations may be reported. In any case, re— liability coefficients are meaningless without some differentiation and definitions3 Brown and others aptly 2 . . 5 Medley and Mltzel, "Measuring Classroom Behavior by Systematic Observation," 250. 53Bob Burton Brown, William Mendenhall, and Robert Beaver, “The Reliability of Observations of Teachers' Classroom Behavior,” The Journal of Experimental Educa— m, XXXVI (Spring, 1968), 3—4. demon their (TPOR ”untr behav judge year, autho 64 demonstrated the necessity of defining reliability in their study of the Teacher Practices Observation Record (TPOR). This instrument was developed for field use by "untrained" observers. Five filmed episodes of teacher— behaviors were observed by a large number of observer judges in different sections of the country. After a year, two of the films were rated a second time. The authors found that: 1. Correlations of observers' total scores within a given film viewing was very good. 2. Correlations of observers‘ scores between film viewings one year apart was poor to fair. 3. Between-observer reliability was .57 or rated fair. 4. Within—observer reliabilities ranging from .48 to .62 were rated fair. 5. Internal consistency reliability was rated good.54 54Ibid., 8. shou] Very coefi coefi the t and 1 emplc crite gardj Effec H4»: l—O—‘v—‘Pfi-‘lhh—(H—‘fn gm n: feet: by S} 65 In selecting an observation schedule, attention should be given to the reliability of the instrument. Very few researchers report more than one reliability coefficient; some do not even define the meaning of the coefficient that is reported. It is easy to see that the highest reliability coefficient could be selected and reported. Several studies already cited in this review employed classroom observations by trained observers as criterion measures. One of many objections raised re- garding systematic observation of overt behaviors is the effect of the observer in the classroom. The objection that teachers and pupils may not behave in exactly the same way when observers are present as they behave when no observer is present has no completely satisfactory answer. The problem of comparing observed and unob— served behavior is akin to that of the small boy who turned out the bedroom light but could never quite make it to his bed before the room got dark. To know how teachers and pupils be— have while they are under observation seems better than to know nothing at all about how teachers and pupils behave.55 Usually, experimenters assume that observer ef— fects are negligible and use all the data collected. Or, 55Medley and Mitzel, ”Measuring Classroom Behavior by Systematic Observation," 248. th! the hor bet tit an< obs see obs twe tha vat fin tea lat ere Var mor 66 they assume that observer effects diminish over time, discard the first few minutes of observations, and use the remaining data. Again, the observer may spend many hours in the classroom so that both teacher and students become accustomed to his presence, resume "normal" ac- tivity, and thus present "true” behaviors when recording is initiated. But this does not solve the dilemma, for observed and unobserved behaviors must be compared to estimate observer effects. No such study has been reported, it seems. However, Masling and Stern hypothesized that, if observer bias diminishes over time, the relationship be— tween the first and final observations should be less than the relationship between the second and final obser— vations, and any other such pairing of intermediate and final observations. Seven trained observers rated teachers and pupils in tWenty—three classrooms. Corre— lations showed no consistent pattern of increase or de— crease over time. The fact may be that teacher and pupil variables under observation occur episodically and are more important than observer influence. Or, the effects tea Tee pre £01 "an bej VeI 196 67 of the observer may be extremely complex and affect vars ious aspects of classroom behavior differentially.56 Though his sample consisted of only ten female teachers, Samph found a definite observer effect when subjects were observed under four experimental conditions. Teachers behame more "indirect" when an observer was present in their classroom whether or not they were in— formed prior to his coming. They used more "praise," "acceptance of student ideas," and less "criticism” when being observed. Samph concluded that the presence of an observer in the classroom leads to change in a teacher's verbal behaviors.57 Summary Some research related to the present study of con- ruency between self—identified attitudes, needs, and eliefs and overt classroom behaviors has been reported. 56Joseph Masling and George Stern, "Effect of the bserver in the Classroom," Journal of Educational Psy— holo , LX (October, 1969), 351—354. 57Thomas Samph, ”Observer Effects on Teacher Be— vior" (unpublished Ph.D. dissertation, University of ichigan, 1968), Dissertation Abstracts, XXIX (February, 69), 2573 A. 68 The review of literature is far from exhaustive, and while support is given to congruency between personality and overt behaviors in some instances, contradictory results are also evident. CHAPTER III PROCEDURES UTILIZED IN THE INVESTIGATION Introduction This chapter provides a detailed description of the procedures followed in this investigation. First of all, a summary of salient characteristics of the study sample is given. Secondly, the instruments used in col— lecting data are described. Finally, the stages of the analysis are summarized, the statistical techniques are discussed and reasons for their use explained. Selection of the Study Population The subjects selected for this study were intern teachers enrolled in the Michigan State University Ele- mentary Intern Program, Spring Term, 1969. This popula— tion was chosen for several reasons. Faculty members associated with the intern program were interested in continued research and willing to cooperate in the 69 bar Pa] Ste Fur Par gat Per Dre r01 7O endeavor. Furthermore, the subjects were beginning teachers; that is, they were completing their first full year of teaching in the elementary school. One hundred seventy—eight out of a total of 191 interns participated in the study. The target population of the investigation in— cluded all elementary education students who participate in a year of internship teaching prior to receiving their bachelor's degree and provisional teacher certification. Participants in the Elementary Intern Program at Michigan State University represented a sample of this population. Furthermore, the 1968-1969 graduates of EIP constituted a sample of past and future Elementary Intern Program Participants. Since subjects included in this investi— Iation were considered a sample of a larger population, >ertinent characteristics were gathered to guide inter— tretation of the results of the present study. Description of the Study Sample One hundred ninety—one intern teachers were en— )lled in the Elementary Intern Program during l968—l969, 19‘ Sdl we] an COT int tee Twc mat COE SSH 146 NO. T tau wit 71 and were completing their teaching internship in public schools of southern Michigan. This figure represents 89.67% of the original 213 sophomores (176 females and 37 males) electing this teacher preparation program in 1967. Not all EIP participants were included in the study sample. Due to ill health of four interns, consultants were not able to complete the classroom observations. In another instance, observation schedules were incorrectly completed and resulted in the loss of six subjects. One interns was disturbed by recorded observations of her teaching; hence, visits to her classroom were discontinued. Two interns entered the program after the personal infor— mation and personality data were gathered. These interns could not be numbered in the study sample. Thus the final sample of this investigation included 37 male (17.42%) and 146 female (82.58%) elementary intern teachers. Deaching Assignments During the 1968-1969 academic year, each intern :aught in a public school district of Michigan affiliated Iith the Elementary Intern Program. Table 3.1 summarizes the whom clas Tabl Teal ASSit Grad Grad Grad Spec. Edllt Tote in m: Whil, grade 72 the teaching assignments of the 178 intern teachers for whom both personality assessments and records of overt classroom behaviors were available. Table 3.l.——Sex and Teaching Assignment of 178 Intern Teachers in the Elementary Intern Program, 1968—1969. Sex , Total Sample Teaching Male Female A531gnment Number Percent Number Percent Number Percent Grades K—2 1 3.23 71 48.30 72 40.45 Grades 3—5 13 41.94 56 38.10 69 38.76 Grades 6—8 14 45.16 13 8.84 27 15.17 Special Education 3 9.67 7 4.76 10 5.62 3 t Total 31 100.00 147 100.00 178 100.00 * Most male subjects accepted teaching assignments .n middle and upper middle elementary grades (87.10%) rhile more female subjects taught in lower and middle 'rades (86.40%). Sul Nur Per Mn 9% Shi Ea Sta 73 Aqe Characteristics The age range of these beginning teachers spans more than thirty years. This is not unusual for the Elementary Intern Program. Since its inception, a number of older married or widowed women enrolled in EIP along with younger female and male candidates.1 The median age in this sample of first year teachers was twenty—two years. Table 3.2 indicates the age distribution of the intern teachers included in the sample. Table 3.2.——Distribution of 178 Intern Teachers by Age. Age Groups Subjects No 21-25 26—30 31—35 36—40 41—45 46—50 50+ Response Number 131 7 ll 12 7 2 2 6 Percent 73.60 3.93 6.18 6.74 3.93 1.12 1.12 3.37 tlmost three—fourths of the interns were in the age cate— Jory. 21—25, which is typical for students who have lBernard R. Corman and Ann G. Olmsted, The Intern— ‘hiE in the preparation of Elementary School Teachers East Lansing, michigan: College of Education, Michigan tate University, 1964), 21- > t t t t conti schor more teaci Cline ment this high coils Table A9 Prior BetWe and After Co] NO re Total 74 continued their college education immediately after high school graduation. About six percent of the interns were more than forty years old at the time of their first teaching assignment. Individuals vary in the time that they are in— clined toward a profession and the time of actual commit— ment to a profession. Most of the subjects included in this study were drawn to the pedagogic profession during high school and committed themselves to teaching during college as Table 3.3 portrays. Table 3.3.—eAge at the Time of Orientation Toward and Commitment to Teaching for the 178 Intern Teachers. Orientation Commitment Age Grouping Number Percent Number Percent Prior to fourteen 61 34.27 9 5.06 Between fourteen and eighteen 67 37.64 58 32.59 ifter entry to college 45 25.28 107 60.11 To reSponse 5 2.81 4 2.25 otal 178 100.00 178 100.01 @939 sch0( port: Howex ceivi tary catic Table Publi Paroc Both Par N0 re Total elect 75 Educational Background In Table 3.4, the type of elementary and secondary schooling for each intern is indicated. The greatest pro— portion of subjects attended public schools at both levels. However, twenty—one percent of the subjects reported re— ceiving some education in private schools at the elemen- tary level and twelve percent reported receiving some edu— cation in private schools at the secondary level. Table 3.4.——Type of School Attended by Intern Teachers during Their Elementary and High School Years. Level Type of School Elementary Secondary Number Percent Number Percent Public 135 75.84 151 84.83 Parochial or private 20 11.24 20 11.24 Both public and parochial or private 18 10.11 2 1.12 to response 5 2.81 5 2.81 otal 178 100.00 178 100.00 Sophomore students from community colleges may lect the Elementary Intern teacher education program. Wm. an 01:] 76 Dne hundred thirty-seven of the 178 subjects in the sample were admitted to EIP from colleges other than Michigan State University. Table 3.5 details the college back- ground of the intern teachers up to the time of their admission to EIP. Fable 3.5.—-College Background of 178 Intern Teachers Prior to Admission to the Elementary Intern Program. College Education Number Percent )nly at Michigan State University 37 20.79 Ip to one year at another college 6 3.37 )ne to two years at another college 87 48.88 ‘wo or more years at another college 44 24.72 '0 response 4 2.25 otal 178 100.01 nly one intern in five had completed all his education at ichigan State University. The remainder had attended iother college for at least part of their education, and fourth had received two or more years training in an— .her institution. 77 Michigan State University College of Education requires a cummulative grade point average of 2.00 to qualify for student teaching. Eighty-eight percent of the intern teachers reported a grade point average equal to or exceeding that minimal requirement at the time of entry to EIP in 1967. Table 3.6 summarizes the statis— tics on grade point average reported by the 178 intern teachers at the time of entry to the intern program. Table 3.6.——Grade Point Average Reported by Intern Teachers at Entry to the Elementary Intern Program, 1967. Grade Point Average Number Percent Less than 2.00 11 6.18 2.00 — 2.49 64 35.96 2.50 — 2.99 56 31.46 3.00 — 3.49 31 17.42 3.50 — 4.00 6 3.37 NO response 10 5 . 62 Potal 178 100.01 )ver half the sample reported a grade point average ex— :eeding 2.50. Just six percent of the subjects did not m 78 eet the minimal g.p.a. requirement for student teaching efore taking their methods courses. ommunity Background Approximately one—fourth (24.16%) of the interns sported that they resided in a city having 10,000 to )0,000 people. More than thirty-three percent of the .terns said they lived in a suburban metropolitan area a city of 100,000 to 500,000 people. Table 3.7 pro— des a complete description of community background for a - 178 intern teachers included in the study sample. ale 3.7.-—Type of Community Background Reported by 178 Intern Teachers, 1968—1969. Type of Community Number Percent ,ropolitan city: over 500,000 people 20 11.24 turban community near metropolitan enter 31 17.42 Y: 100,000—500,000 people 29 16.29 urban community near a city 9 5.06 Y of 10,000 to 100,000 people 43 24.16 1 Of 2,500 to 10,000 people 25 14.04 il community of less than 2,500 BOple or a farm 17 9~55 TeSponse 4 2-25 ll 178 100.01 t0] _— e a w t. ti mind A _Dg m bit in att rel Lee 0 \ Cal Yer tla tud Min 79 Instrumentation Several standardized measurement devices were used this research: The Minnesota Teacher Attitude Inven~ gy, nggrds Personal Preference Schedule, the Rokeach gmatism Scale (short form), and the Observation Schedule d Record (OScAR). The purpose, validity, and relia~ lity of measurement for each instrument is summarized this section. e Minnesota Teacher titude Inventory The Minnesota Teacher Attitude Inventory (MTAI) s deve10ped by Cook, Leeds, and Callis to assess teacher :itudes which would predict the type of teacher—pupil Lations a teacher would maintain in the classroom.2 ads originated the work in his doctoral research devoted . 3 the develOpment of the Teacher—Pupil Inventory. He 2Walter W. Cook, Carroll H. Leeds, and Robert lis, Minnesota Teacher Attitude Inventory: Manual (New k: The Psychological Corporation, 1951). 3Carroll H. Leeds, "The Construction and Differen— 1 Value of a Scale for Determining Teacher—Pupil Atti— ES” (unpublished doctoral dissertation, University of nesota, Minneapolis, Minnesota, 1946). r1 .1 m tea COD agr ind SCO 80 first identified desirable and undesirable characteristics of teacher—pupil relations. These extreme categories guided Leeds in selecting items regarding: (l) the moral status of children; (2) discipline in the school and methods of dealing with discipline problems; (3) prin— ciples of child development and behavior; (4) principles of education related to philosophy, curriculum, and ad— ministration; and (5) personal likes and dislikes of the teacher.4 The revised version of the attitude inventory contains 150 statements to which the subject reSponds on a five-category Likert—type scale ranging from strongly agree to strongly disagree. Though the responses of an individual are considered neither right nor wrong, a scoring key'using the common labels is provided. The iighest possible score is +150 and the lowest is —150. 1eachers who rank high are assumed to have the ability .0 maintain a more harmonious relationship with children, 0 be more permissive, to create a cooperative learning nvironment, and to have fewer disciplinary problems. Cook, Leeds, and Callis, Minnesota Teacher Atti— ude Inventory: Manual, 10. in. Tht tez te; Put 81 Low scoring subjects are thought to be strict discipli~ narians, content—oriented rather than child-centered and less open to a variety of approaches to problem—solving.5 The MTAI was devised to discriminate sharply between teachers who maintain extremes in their rapport with children.6 Leeds correlated MTAI scores of one hundred teachers with ratings by principals, an eXpert, and pupils. Each respective rating correlated .46, .59, and .31 with the Minnesota Teacher Attitude Inventory score. Using combined multiple weights for the three validating criteria, an overall .63 correlation was obtained.7 Thus, the MTAI seemed to have some predictive power regarding teacher—pupil rapport when compared with ratings by prin- cipals, pupils, or an expert. However, it appeared that 5Robert G. Oana, ”An Analysis of the Use of the Minnesota Teacher Attitude Inventory in a Preservice Pro— gram in Childhood Education" (unpublished Ed.D. disserta— tion, Columbia University, 1965), 32-33. 6 . . Cook, Leeds, and Callis, Minnesota Teacher Atti— ude Inventor : Manual, 4. 7Carroll H. Leeds, "A Second Validity Study of he Minnesota Teacher Attitude Inventory,” Elementary chool Journal, LII (March, 1952), 396e405. t i .I. 1. l\ 1. l a t R II) s C .L C E l v A R 0 82 idging by an expert was a better predictor than either E the other ratings. Using very similar procedures with a slightly >dified version of the inventory, Callis correlated NEAI id ratings scores of seventy—seven teachers. Correla— Lon between principals' ratings and MTAI scores (r = .19) as much lower than that found in Leeds' sample (r = .46). itings by two experts correlated .40 with the NEAI scores :ompared to .59 for Leeds' ratings of teachers in his ample). On the other hand, Callis reported a much higher )rrelation between pupils" ratings of their teachers and 1 1e MTAI score of those teachers than did Leeds (for i llis' sample, r = .49; for Leeds' sample, r = .31). 8 ther the ratings method of gathering evidence of val— ity is unstable, or the instrument itself has low lidity. In his factorial study of the Minnesota Teacher titude Inventory, Ferguson reported that only one type 8Robert Callis, "The Efficiency of the Minnesota icher Attitude Inventory for Predicting Interpersonal Lations in the Classroom," Journal of Applied Psychol— D XXXVII (April, 1953), 82—85. of at mute drawn of th than inven time ; towar< studer dent t brief Who er Points hOWEVe OVer t in mea \ Minnes 35333 Teachei Igésigfl 83 of attitude is measured by the NEAI.9 Considering the number of items and the five sources from which they were drawn, this is rather surprising. It justifies the use of the total MTAI score as a variable in itself rather than considering responses to subsets of items in the inventory, however. Studies of the consistency of measurement over time seem to indicate that respondents' attitudes shift toward "a more realistic View” as students complete their student teaching. Day found that average scores for stu— dent teachers showed a mean loss of four points in that brief period of time. Test—retest differences for seniors who entered teaching showed an average loss of twenty points after a six—month period. Non—teacher graduates, however, only lost an average of one and one—half points over the same period of time.10 Oana reported an increase Ln mean MTAI score for sixty—four participants in student 9John L. Ferguson, Jr., "A Factorial Study of the linnesota Teacher Attitude Inventory" (unpublished Ed.D. lissertation, University of Missouri, 1953), Dissertation bstracts, XIII (No. 6, 1953), 1087. 10 Harry P. Day, "Attitude Changes of Beginning achers After Initial Teaching Experience,” Journal of acher Education, X (September, 1959), 326—328. If? Ira. WE M1 teaching. However, the gains were largely negated by a definite drop in scores after one or two years of teaching experience. The Edward Personal Preference Schedule Murray‘s taxonomy of needs was briefly reviewed in Chapter II. It seemed appropriate to base the discussion of personal manifest needs on Murray's theory since Ed— wards studied with Murray. Edwards extended the work of Murray by defining the manner in which each inner need would manifest itself in the overt behavior of an indi- idual. He then constructed an inventory to assess these anifest needs. The Edwards Personal Preference Schedule (EPPS) is 225—item ipsative measuring device designed to provide ithin individual comparisons of the relative strength of ifteen "manifest needs”: (1) Achievement; (2) Deference; 3) Order; (4) Exhibition; (5) Autonomy; (6) Affiliation; ll . . Oana, ”An AnalySis of the Use of the Minnesota acher Attitude Inventory in a Preservice Program in ildhood Education.” the Cit W66 art st: abc m Scr and (7) Intraception; (8) Succorance; (9) Dominance; (10) Abasement; (ll) Nurturance; (12) Change; (13) Endurance; (l4) Heterosexuality; (15) Aggression.12 Edwards reported that split half internal consistency reliability coeffi~ oients ranged from .60 to .87 with a median of .78. One— veek retest reliability coefficients ranged from .70 to .87 with a median reliability of .83.13 Early norms for the EPPS were based on responses if 1,509 liberal arts students. In the revised version if the EPPS, extensive adult norms were also provided. eachers' needs profiles compared more favorably with ollege educated adults of the same age than with liberal ts students' profiles.l4 In order to assess the extent to which three in- ruments validly measure the fifteen needs mentioned ove, Poe intercorrelated the fifteen needs across the 12A. L. Edwards, Edwards Personal Preference hedule: Manual (Rev. Ed., New York: The Psychological rporation, 1959). The fifteen need variables are de— ribed in Appendix A. 1 3Ibid., p. 19. l 4Getzels and Jackson, ”The Teacher's Personality Characteristics," 546. inventories (one of which was the EPPS). A multitrait- .ultimethod matrix was constructed from these correla— ions. A11 correlations met the criterion for convergent alidity; that is, the correlation of different measures .f the same trait were significantly different from zero =nd sufficiently large to warrant further eXploration. he criterion for discriminant validation was also met. oe concluded that each of the three methods measured :he needs and that the fifteen needs were meaningful var— .ables which may be related to other factors of interest . 15 0 educational researchers. he Dogmatism Scale The Rokeach Dogmatism Scale measures individual ifferences in openness or closedness of belief systems, aneral authoritarianism, and general tolerance. Items alected for the scale are familiar to the average person l everyday life, and transcend specific ideological posi~ ons. Form E, the most widely used version, consists of 15Charles A. Poe, "Convergent and Discriminant lidation of Measures of Personal Needs," Journal of Edu— tional Measurement, LXI (Summer, 1969), 103—107. for rel men scc tot obt ran pro sit reg tes hal Spo tot; Sew the SCo: mat; 87 forty statements to which the respondent indicates the relative strength of his agreement (+1 to +3) or disagree— ment (—1 to —3). For all statements, agreement is scored as closed and disagreement as open. Thus, the higher the score, the more dogmatic an individual is said to be. The total score on the Dogmatism Scale is the sum of scores obtained on all items.16 Rokeach reported reliabilities ranging from .68 to .93 for the forty—item scale.l7 Length of measuring devices consistently poses a >roblem for researchers who must collect data in field iituations. Troldahl and Powell, to solve this problem 'egarding the Dogmatism Scale, administered the forty—item est to two samples of adult subjects from which a split— alf reliability coefficient of .84 was obtained. Re— pondents’ scores on each item were correlated with their otal score to obtain a homogeneity index for each item. everal short forms of the scale were constructed using 1e items having the highest item—total score correlation. :ores on each short form were then correlated with dog— .tism scores on the complete scale. The twenty—item l6Rokeach, The Open and Closed Mind, 71—73. l7Ibid., 9o. [on IE?“ 88 version (just half as long as the original Dogmatism Scale) correlated .95 and .94 for the Boston and Lansing samples respectively. According to these figures, the twenty—item short form is a good predictor of what a re— . 18 ' . spondent would obtain on Form E. The twenty—item ver— . . . . . l9 Sion was administered in this study. Jay questioned Rokeach's assumption of the uni— tary basis for the belief—disbelief system. He analyzed the response patterns of twenty—nine college subjects by means of Q—technique factor analysis and rotated three factors to simple structure. Three factors described the students: (1) Open—minded, tolerant, nondogmatists; (2) true believers who have a profound and generalized Eear of life; and (3) true believers who are authoritar- .an.20 Each factor found by Jay seems to correspond with 18Verling C. Troldahl and Fredric A. Powell, "A hort-Form Dogmatism Scale for Use in Field Studies," ocial Forces, XLIX (December, 1965), 211—214. 19 . . . The twenty—item scale is reproduced in Ap— endix B. 20Rutledge L. Jay, ”Q—Technique Factor Analysis E the Rokeach Dogmatism Scale," Educational and Psycho— )gical Measurement, XXIX (Summer, 1969), 453—459. in. tie and Gid vid be i 91 a [a Is tail to I elem the 194. Metk Edut -~. 1964 89 a particular dimension of Rokeach's definition of dogma— . 2 tism, however. 1 From their investigation of the scoring methods and construct validity of the Dogmatism Scale, Korn and Giddan concluded that response set may occur in an indi— . 22 . Vidual's manner of response to the statements. This is not an unusual phenomenon, though, and may in itself be an indication of the person's open—closed—mindedness. The Observation Schedule ___________.___________ and Record In addition to the three inventories used to ob— tain indirect evidence of attitudes, beliefs, and needs 3f intern teachers, a standardized instrument was employed :0 record overt behaviors of intern teachers in their elementary school classrooms. Medley and Mitzel developed :he observation form, OScAR (Observation Schedule and 21Rokeach, "The Nature and Meaning of Dogmatism,” 94. 22Harold A. Korn and Norman S. Giddan, "Scoring ethods and Construct validity of the Dogmatism Scale, ducational and P5 cholo ical Measurement, XXIV (Winter, 1 964). 867—874. R. g h a d i a .1 C p g .0. 0.. LL 0. b s .N: m\U] U J\3.Av 9O ecord), for use in a follow—up study of teacher education :aduates. By modifying and combining items constructed { Cornell, Lindvall, and Saupe,23 and Withall24 the 1thors (1) increased observer accuracy by reducing the .fficulty of judgments on the part of the observer; 5) employed but one observer per classroom; and (3) sep— 'ated the process of observing from the process of scor— 19. The observer merely checked a behavior which oc— rred without judging its possible significance. Items related to teacher—pupil relations, inde— ndent teacher and pupil activities, affective behaviors ouping arrangements, materials used and subject matter served were recorded in a five—minute observation :iod. In the second five minutes the observer attended verbal behaviors of the teacher. The observer alter— :ed observation of general class activities with verbal laviors for a thirty—minute period. I 23F. G. Cornell, C. M. Lindvall, and J. L. Saupe, Exploratory Measurement of Individualities of Schools Classrooms (Urbana: Bureau of Educational Research, versity of Illinois, 1952). 4John Withall, ”Development of a Technique for Measurement of Socio-Emotional Climate in Classrooms,” rnal of Experimental Education, XVII (March, 1949), —361. 91 Data collected in the original study of forty-nine :ginning teachers were analyzed in several steps. First, ,ems were combined into twenty ”keys" and scored. Mean :ores on each key were studied to explore possible reli- ile differences for the forty—nine classrooms; six keys re discarded as unreliable. A factor analysis of the maining fourteen keys revealed three dimensions corre— onding to the first three rotated factors: Emotional imate, Verbal Emphasis, and Social Structure. Emotional imate referred to the amount of hostility observable in classroom. A high score indicated external manifesta— on of warmth and friendliness with rare occasions of stile reactions. Verbal Emphasis indicated the degree which such verbal activities as reading, writing, or a of the textbook predominated. Social Structure.re— tred to the amount of pupil autonomy as opposed to 'uctured group activities. High scores on this dimen— »n indicated that the class functioned quite autono— sly with little teacher talk. Reliability coefficients estimated the correlation Ween the mean of all the scores assigned to the :hers by the six observers in twelve observation F": 92 periods and means of scores that would be assigned to the same teachers by six different observers visiting each teacher at tWelve different times. Table 3.8 shows the reliability and the intercorrelations between each of the three dimensions described above.25 Table 3.8.-—Intercorrelations Among Three Factor Scales Based on OScAR Scores Of Forty—Nine Beginning Teachers.* Scale EC VE SS Emotional Climate (EC) (.903) —.004 —.110 Verbal Emphasis (VE) (.770) +.028 Social Structure (SS) (.826) Reliabilities are reported in the diagonal. Practically every category of OScAR was used in this re— a search.26 Medley's recent revision of the verbal section )r K—scale (now called OScAR 5V) was not used since 25Donald M. Medley and Harold E. Mitzel, ”A Tech— lique for Measuring Classroom Behavior," Journal of Educa— ;ional Psychology, XLIX (April, 1958), 90. 26 Medley and Mitzel, "Measuring Classroom Behavior Y Systematic Observation," 278—280, provides a copy of ScAR 2a. A sample observation form and description of he categories used in this investigation appears in Ap— endix C. servers could not be sufficiently trained in the time ailable for that phase of the study. OScAR III does not probe aspects of classroom havior related to pupil achievement of cognitive ob— ctives. The dimensions that it does measure probably dicate the most obvious differences among elementary asses. Reliable measurement of obvious differences is latively easy; to measure more subtle and perhaps cru— al differences is more difficult though not impossible. Procedures of the Investigation Several steps were involved in the study proce— 7es. Data collection included two distinct phases: ipletion of personal information and the three per— .ality inventories by the 178 subjects at entrance to Elementary Intern Program, and six separate observa— ns of overt classroom behaviors of the intern teachers ing the final weeks of their internship teaching, May— a, 1969. The latter phase occurred approximately 27Ibid., 286. 27 pri Ter u d\ E r 0‘ t Rok en (301 Spe sYs aCc ’_ 94 -l/2 years after entrance to EIP. Data were prepared or analysis on the CDC 3600 computer at Michigan State niversity. Collection of data, preparation for analysis, nd subsequent analysis are described in detail in this action. llection of Personal and rsonalit Data During the first week of their methods courses and ior to contact with children in elementary school class— oms, the EIP pre—interns for Fall Term, 1967, and Winter rm, 1968, completed the following forms: the Teacher ucation Inventory, the Minnesota Teacher Attitude Inven— gy, the Edwards Personal Preference Schedule, and the :each Dogmatism Scale. Each of these paper—pencil in— itories was completed under the supervision of the EIP iter director at ten off—campus locations. Forms were .lected, analyzed, and results for each recorded. A cially designed computer—based information retrieval tem was developed to maintain records and provide ready ess to data for further analyses. 95 Collection of Data on Overt :lassroom Behaviors In order to relate self perceptions of intern :eachers to their overt teaching behaviors,-observations .n their classrooms were required. Intern consultants egularly visited and worked in interns' classrooms. It as assumed that consultants' presence in the classroom ould not offset the general learning climate. Thus irty—seven consultants could be asked to make six sep— rate observations in each of their intern's rooms, and 3 record overt behaviors using the Observation Schedule id Record. On March 6, 1969, the researcher attended 1e EIP center directors'regular staff conference. The asearch project, procedures for training the observers, id making the observations were discussed and approved ‘the directors._ aining the Observers A biannual conference for student teaching coor— mators, EIP directors, and intern consultants is held‘r MSU's Kellogg Biological Station, Gull Lake. At the am 96 pring Conference, April 23—25, 1969, four hours were svoted to study of the Observation Schedule and Record )ScAR) and practice in recording behaviors observed in ilmed teaching episodes. To facilitate the study, con— :rence participants received the OScAR Manual and re— irding form a week prior to the conference. Training observers required several steps. First, detailed study of the one hundred categories was made. terpretation of the categories was facilitated through ference to descriptions included in the manual. After viewing the categories, the format of and procedures for ing the recording form were presented to the conference :ticipants. The recording sheet included one hundred :egories for general classroom activities which were 'anged for ease of marking. The categories were grouped ler several headings: Teacher and Pupil Activities, uping Arrangements, Teacher and Pupil Materials and ject Matter Observed. A twenty—five minute observation iod consisted of the following segments: 8See Appendix C for a copy of the OScAR Manual recording form. 97 Five—minute interval Observer Action First Check observed behaviors in column labeled I. Second Observe only. Third Check observed behaviors in column labeled III. Fourth Observe only. Fifth Check observed behaviors in column labeled V. Once the conference participants were familiar th the categories and the procedures for recording, Lmed classroom episodes were used to acquire facility recording overt behaviors. Observers checked as many art behaviors as they observed during the filmed epi— les; subsequently a comparison of recorded behaviors : made. This pattern of observing—recording, then cussing—comparing was followed throughout the training sion. Films used included SRA's short segments from Teaching Problems Laboratory, numbers 29, 8, 18, 10, 1:29 and several 16 mm. films approximately thirty JteS long: (1) Keith, A Second Grader; (2) Dick, A 29Donald Cruickschank, et a1., Teaching Problems >ratory, Critical Teaching Problems (New York: Science aarch Associates, Inc.). If 98 Fifth Grader; (3) Greg, An Eighth Grader; (4) Unit Teach— ing in Kindergarten; (5) Unit Teaching in Fourth Grade Science; and (6) General Objectives. Not all the avail— able films were needed during the training sessions. Before leaving the conference on April 25, 1969, :ach intern consultant was provided a stopwatch and re— ording forms to be used in observing teaching behaviors f each intern teacher in six separate sessions. Detailed irections were given and any remaining questions were nswered. Five regional meetings were held with intern con— iltants during the two weeks following training at Gull ike. Dates of these conferences and the EIP Centers >ncerned were: April 29, 1969 - Alpena April 30, 1969 — Bay City—Saginaw and Port Huron at Bay City May 2, 1969 — Grand Rapids May 5, 1969 — Detroit, Macomb, and Pontiac at Macomb Center; East Lansing at East Lansing May 9, 1969 — Battle Creek :ther questions of intern consultants were answered and 1sistency in interpreting categories and recording 99 ahaviors was checked. Actual teaching behaviors and leed episodes were observed and recorded to check con— _stency. gpgration of Data r Analysis Following the regional training conferences, each the thirty—seven consultants scheduled six observations each of the interns they supervised. These observa— >ns were completed and returned to the researcher by 1e 15, 1969. Thus all observations were made in a six— ek period near the end of the subjects' first year of Ching. In all, 1,068 observations (6 x 178) were made. y were processed in three stages. 1. Each behavior in the separate categories was summed across the three time intervals of an ob— servation. 2. These totals on each observation form were summed across the six observations for a particular in— tern teacher. Where fewer than six observations had been made, one of the remaining observations 100 was randomly selected and each recorded variable duplicated. 3. The raw score for each of the one hundred cate- gories was recorded, verified, key—punched, and verified again. Raw scores for a particular be~ havior of a given subject ranged from zero (never observed in eighteen five—minute periods) to eighteen (observed in every five—minute period). sis of Data Data on the personality variables and the overt room teaching behaviors of the interns were subjected :tor analysis. Preparation of the data, statistics, :er routines, and the results of each step are sum— :d in Tables 3.9 and 3.10. The statistical tech— , employed are described in detail subsequent to mmaries. 101 mosMHMm> Hopou mfiu mo.mmm How one lussooom GOMDSHom poumuou smuflm ens so mHODUMM 039 .manmesm> Sumo Mom memouufix.©:m mmwGSoxm .GOHHMH>®© Cuspcmum .cmoz .»UHmHo>. IHQD.ou89m advances .QOHumo Ifisssaoo mo unofiuummoo o3» mo moos>som noummmom .msospmnom Hmcomoavuo can mpqosomaoo HmmHquHm .mflmmamam Moscow hpflmuo>fiso ounpm :mmafloaz .m Honfisz aoflpmauomoa mmanmm seem .eoeumum ucofiflnomxm HMHDpHSOHHmd Asasmmv moaumepmpm oammm .muonommp finesse uSmHoIMpano or» we moHoom smaumsaoo map and Hess map .mouoom mmmm zoouHSOM on“ mo Aonoom mo noncommos maoam mofisflumaflaflm mcflu>amcm Mom oswflssooulov mflmhamsm Houomm .oQHuDOH mflmwamcm Houomm osu mo macapmuflsfla was poms 0p mamamm Hmuou esp scum whosomou canvas “swam Imunmflo mo Gosuooaom Hoocmm .quflfihopop on son pH503.oH5mMoE o>fiuMmmH mmmm map so whoom Hmcfim onu.pm:p 0m mammamqm use EOHM whoom unmaommnm mo deflwmflfio .mHmnomou ssouss mud mo mmHoom Emflpmamom as» new Haas map .mmmm was mo moaanHm> sooprm map How moflpmapmum Damon Ho coasmasoamo lIlIIIIIlIIIIIII|IIIIIIIIIIIIIIIIIIIIIIII mpHSmom ocflusom .moHQMHHm> mpflamqomuom mo mHmMHmcm as mommu meMqum Mll.m.m GHQMB 102 .mwmhu muHHmcomHom 03.... omens 93:0on gonna mha on“ mo sowpmummom .mflmhamcm Moscow map as poms moHQMHHm> huwamcomnom noouxflm on“ NO sumo Mom musmflos upon pad musoHOHmmooo sonmonom .muouomm oBu on“ no moHQMHHm> muflamqomnom coopxflm may mo some now mouoom puppcmum pouebfioz .psmuHSmsoo mosumflpmum.m .Mommuo 5:06 >9 soupflss Emumoum Hafioomm MpHmH0>HaD mumpm cmmflnoflz .n Hwnfidz COHumfiHomwQ mmaumm seem .eoeumum uqoafluomxm HMSMuHSOHum¢ Amoumswm pmmoqv mu , .NH nonesz moflhm Emumosm .sOfiHMOHQSEEOO mo namsunmmwo .m mama .mwcowoflmmooo.a0fimmoum now no memos one co.mHo£omop quosqfl wed was mo some Mom wmmu hpflHMGOmuom mo GOHDOHUme .b .poumamcm Houomm soon was roars mwenmenm> spasm Isomuom coouxflm may no mpoon lawn psmflolhusmwm mo mouoom paw mommu muHHMGOmHom osu soozpon mmfianOflumaon mumsflpmo ou mosmswm pmmoq.wo vogue: was ocwms QOmeonom Hmoqfiq .o .muouomw 03» we“ so muoonnsm snmwolhunmflo may mo Soto Mom mmsfipmoa Houomw was mo mflmmamsm .m mudsmom oaflusom mammamss . .mmmosuomwn map How oofidoamacmflm mo mumps .mHosomwp Guqufl mba How muouomm o>Hm 03# «0 some so moHoom uouomm .musofisoufl>co waflsumoa o>Hm onflnomop on poms mums muopomM 0>Hm muons .oquHso> HopOp map mo mmm Ham.msflpssoo 10m muouomm o>am potaofl> GosBSHom popmuos cussom .maoausaom popmwos pfimflm 103 .msoamfiooqfi mums sump ouoss poosoooh mos magma Ism> M MD ado: .pmNMHmcm moanmaum> mes mo some Mom mamouusx was mmonaoxm .Goflumfi>wp Cuspcmum eddoz ounaum>fipasz can opossm> IHGD ”oosmflum>flpH52: How EMHmOHm ZZHm .GGHm Jo afiouwh .NH moHHm Emumoum .QOHUMUHGSE IEOU mo pawfiuummwo .N mdmz .hpflmuo>flso opmwm gmaflsoflz .qOHBGOflQDEEOO mo pcofiphmmoo onu mo move I>Hom goumomom .mGOMUMBOM Hmsomonuho was mucosomfioo Hmmfloaflnm .mflmwamcm Hepomm .s Honfisz “Homom HMDflQSUoE Soumomom ooqofiom Hafioom Mom ousseumsH Housmaoo «Emsmoum cosumaouuoo memo mpoamaoosH .mmoomH osflusom "so mHoSUMou seduce wba How mH0p0o> qmofi mo mouse Inm> mo mHmMHmsm opmflsm>muadz .GOHpDHOm poumuon sunsom was Mom mouoom Houomm mo cosponsmfiou .mHosomop GHQHGH mha mo sumo How moanmflum> touted: use map mo Aommulmv mflmhamqm Honowm .mHoaumou unease whH Mom moHQmHHm> soawm>Hmmno Eooummmao ponpsss one men Mom moflumfipoum Damon mo coeumasowmu .H mpHSmmm .meQMHHm> H0H>M£mm Eooummmav uHo>O commas meandom mflmwamsm m ado mo mammamed ea museumnl.oe.m deems 104 .mHo>oH osflnomwu mSOHMm> mo mcomflummfioo o>ao3b How wands as moocmsowwflp mo oocmoHMHsmHm mo memos waspsom zsz .HHoo some cflaufla mwuoom mo soap Imfl>op UHMwGMUm was new: wsfiusom BmBmm .muflmno>fisb wumum.emmanoes .eoepmosem mo omoaaoo .MpSDm poosd>p¢ now H00£om men «0 moos>nom nonmom umm .some .sms =.6oemenm> IOO_pcm ooqmfium> mo mflmmamss .mouoom uqmsfiafiuo Imam pamoflwflsmfim mo mpom 03» so mHo>oH mcflaomop new names coospon moosoquMHp mo oocmoflMHcme umop as _mcomfiummfioo 003 pmom .mmmwflom .muonowou seep Imogen Hmaommm 6am .muo .msm .NIM imoamfiom .monE "msoofl Insm wee mo mouoow psmsflssso Imam pew museum Houomw How moauwflumum Demon mo coaumusdaou mHo>oH mafiaomop xow momwu mpfiamsomuwm .0 .n .w .m muaswwm oqfludom mflmhamsfl 105 Factor Analysis Factor analysis is a multivariate analytical tech— ue in which many variables are observed simultaneously hout establishing priorities among them. In education, e of the most important variables cannot be controlled the laboratory, and must be examined in_§ipg. In such es, many variables are recorded to reduce the possible rces of error.30 When the numerous observed variables intercorrelated, a matrix results. This intercorrela- n matrix may be factor analyzed to serve any or all of following goals: 1. To parsimoniously describe n variables in k factors where k is less than n; 2. To suggest hypotheses; or 3. To support or disprove hypotheses related to structure or organization. 30Raymond B. Cattell, "The Three Basic Factor— ytic Research Designs——Their Interrelations and De— tives," Psychological Bulletin, XLIX (September, ), 514. 106 h factor obtained in the factor analysis is a condensed tement of linear relations derived from the set of var— les.31 Factor analysis, from another point of View, de— mines the variance components of the total common :or variance. Several sources of variation contribute he total variance of a measure: a - Vt is the total variance of a measure; V o is the on factor variance; V equals the specific variance; P error variance of the measure is denoted Ve. The er of factors obtained in describing the common var— : of the variables indicates their complexity. A .re may be factorially ”pure" or ”complex,” saturated . 32 only one factor or loaded With several. In a factor matrix, correlation coefficients or loadings) eXpress relations between the variables 31H. J. Eysenck, "The Logical Basis of Factor Sis.” American Psychologist, VIII (March, 1953), )8. 32Fred N. Kerlinger, Foundations of Behavioral fch (Chicago: Holt, Rinehart and Winston, Inc., 655-656. 107 .alyzed and the underlying factors. In most factor alyses, the factor matrix is submitted to either ortho— nal or oblique rotations to facilitate interpretation the factor structure. Thurstone established simple . . 33 ructure prinCiples for analysts to follow. These . . 34 les apply to both orthogonal and oblique solutions. Jure 3.1 portrays a generalized matrix showing a three— :tor solution. Variable Xl correlates r with the 11 "st factor and rl3 with the third. The last column of . 2 . : table is called the communality or h . This "common tor variance” is determined by summing the squares of factor loadings for a particular variable, such as: 2 2 2 2 2 ) + + (rlk) re k is the last factor extracted from the matrix. 33Ibid., 667—670. 34Rotation consists of a mathematical view of from various angels to simplify the factor structure .dentify the most invariant factor structures. Ortho— rl rotation maintains independence between two factors hat their correlation is zero. Oblique rotations do place this constraint upon the factors. 108 Factors riables —————————_—________________ h2 r 11 r12 r13 hX 1 r21 r22 r23 hX 2 r31 r32 r33 hX 3 2 rnl rn2 rn3 hX n ure 3.1.——Generalized Factor Matrix. The researcher may select one of several types of :or analysis, depending on the kind of variables com— - . . 35 ting the correlation matrix. In R—type factor anal— ,, variables or tests for a sample of people are corre— d as in the analysis of OScAR III variables. If, how- , Similarities between persons interests the 35Catell, ”Three Basic Factor-Analytic Research gns——Their Interrelations and Derivatives," 499—520. 31m S. MacLean, Jr., "Some Multivariate Designs for Jnications Research," Journalism Quarterly, XLII imn, 1965), 614—622. 109 :stigator then people are correlated with each other a sample of responses or tests. Q-technique, as this s of factor analysis is called, was utilized with the .ty—eight subjects' personality test scores on sixteen ables. Procedures in using R and Q types are de— bed in the following paragraphs. pe Factor Analysis R analysis involves correlating and factoring ables for a sample of persons with time held con— _ 36 n The principal aspects of factor analysis of >ns' overt behaviors is illustrated in the following . which parallels this investigation. First, a score matrix for 178 intern teachers e hundred dimensions of classroom behaviors was lated. 36MacLean, "Some Multivariate Designs for Com— ations Research," 614. 110 Classroom Observation Variables rson X1 X2 X3 X4 X5 ' X100 11 6 0 3 4 . . . O 11 15 9 0 2 . . . O 16 8 0 2 9 . . . 0 12 15 5 0 1 . . . O are 3.2.——Score Matrix of One Hundred Classroom Obser— vation Variables for 178 Subjects. son 3 was observed questioning pupils more often (Var- -e 1) than Persons 1, 2, or 178. Every variable in score matrix was correlated with every other variable, ding a correlation matrix of the following form: X . . . ables X1 X2 X3 X4 5 x100 1.00 .28 1.00 26 .07 1.00 .14 25 .22 1.00 —.09 —.20 .03 .27 1.00 —.04 .08 —.02 .OO —.06 . . . 1.00 'e 3.3.—-Correlation Matrix of One Hundred Classroom Observation Variables for 178 Intern Teachers. 111 Lriable 1, "Teacher questions, pupil answers," correlated :8 with Variable 2, "Teacher answers pupil question," and .6 with Variable 3, ”Teacher interrupts pupil." Variable ‘0 was checked if some subject other than those listed d been observed. It had very low correlations with all her variables. The correlation matrix was analyzed by means of thogonal rotations to extract the underlying dimensions Dng the one hundred variables. Eight rotations were ie, each successive rotation extracting an additional :tor. Table 3.11 summarizes the proportion of variance >lained by each factor in each rotation. )1e 3.11.——Percent of Variance Explained by Each Factor in the Eight Rotated Solutions of 100 Class— room Observation Variables. Factors Total ation Variance I II III IV V VI VII VIII IX Explained st 8 6 14 ond 7 6 5 18 rd 5 5 7 5 22 rth 7 5 5 4 4 25 th 5 7 5 4 4 4 29 t 7 5 5 4 4 4 4 33 anth 5 7 5 4 4 4 4 3 36 1th 6 5 5 3 4 4 4 3 4 38 112 It can readily be seen that additional factors explained very small amounts of the total variance. If communalities rather than unities had been inserted in the diagonal of the matrix, more true variance could have seen explained. Though the eighth rotated solution ac— :ounted for the greatest amount of variance (38%), the )verall description of data would not be as parsimonious is a solution having feWer factors. Cattell suggests a scree test for selecting the minimum number of factors to be extracted from the matrix. or this test, the eigenvalue curve is plotted.37 The cree point or inflection in the curve indicates the umber of factors accounting for most of the variance. fter the scree point is reached, the curve tapers to a elatively straight line or scree indicating that succes— ive factors explain equally small amounts of the total ariance. Figure 3.4 shows the eigenvalue curve and the :ree point for classroom observation data analyzed in 37Raymond B. Cattell, ”The Meaning and strategic :e of Factor Analysis," Handbook of Multivariate Experi— ental Psychology, edited by Raymond B. Cattell (Chicago: .nd McNally & Co., 1966), 206. Eigenvalues are latent tots in the solution of a matrix form of an equation. 113 :his study. Thirty—two eigenvalues exceeded the threshold Ialue of one. [—1 O I ize of atent 6- Roots Scree Point 1 2 3 4 5 6 7 8 9 '10 11 12 Successive Factors Extracted Lgure 3.4.——Scree Slope Indicating the Minimum Number of Factors to be Extracted from the Factor Matrix Based on 100 Overt Classroom Behaviors. Le fourth rotated solution was selected for further _alysis on the basis of: l. The scree test; 2. The proportion of variance explained; 3. Parsimoniousness in describing the variables; and 4. Conceptual meaningfulness of the factors. 114 By analyzing the factor loadings, five learning environ— ments were identified and described: Supportive, Conven— tional, Independent, Enriched, and Hostile. Descriptions are presented in Chapter IV. Qrtype Factor Analysis Q analysis correlates and factors persons for some sample of tests, again holding time constant. According to Cattell, Q analysis is simply the obverse of R—type analysis. Q technique is concerned with the patterns of response of an individual and the degree of similarity betWeen persons in their patterns of response. Correla- :ion and factoring places an individual with other persons vho respond as he does. While R—type analysis is norma— :ive, Q—type is ipsative. Eighty—eight subjects were randomly selected from Lhe total sample. The matrix of personality scores for these individuals was tranSposed so that columns contained ixteen personality scores for one individual and rows 3 . . . 8MacLean, "Some Multivariate DeSigns for Commun- cations Research,” 614—616. 115 ontained all personality scores from the same scale or est. Figure 3.5 indicates the form of the matrix and ome of the score values. arsonality Person score 1 2 3 4 5 88 :hievement 15 16 19 14 15 . . . 15 sference 15 10 20 12 6 . . . 9 rder 15 11 10 8 11 . . . 3 ngatism 6O 80 66 64 58 . . . 49 igure 3.5.——Transposed Matrix of Personality Scores for 178 Elementary Intern Teachers. Each person was then correlated with every other arson, yielding an 88 X 88 matrix of correlations as >1lows: rson 1 2 3 4 5 . . 88 l 1.00 2 0.95 1.00 3 0.85 0.82 1.00 4 0.78 0.81 0.68 1.00 5 0. 92 0.92 0.86 0.79 1.00 8 0:88 0.89 0.92 0.78 0.88 . . . 1.00 Jure 3.6.-—Intercorrelation Matrix of Eighty-Eight In- tern Teachers. 116 arsons 1 and 2 correlated .95 while Persons 3 and 4 cor— alated only .68. The first pair responded similarly on me sixteen tests and the latter pair differed markedly. Two and three factor solutions were obtained, :counting for 92 and 94 percent of the total variance aspectively. Figure 3.7 shows the eigenvalue curve for me factor solutions. Only four latent roots exceeded 1e threshold value of one. 80— 70— . f 60— Lze o 50_ atent 40_ loots 30— 20— 10— 1 2 3 4 Successive Factors Extracted gure 3.7.——Scree Slope Indicating the Minimum Number of Factors to be Extracted from the Factor Matrix Based on the Personality Data. viously, the third factor contributed little to either :simoniousness of description or proportion of variance )lained. Thus the first rotation, i.e., the two—factor .ution was selected for further analysis. 117 Each person was assigned to a particular person— Llity type on the basis of his highest loading. To de— Lermine the "personality profile" or describe the ”type," 'eights derived from the factor loadings were applied to he original score matrix for the eighty-eight subjects. he distributions were standardized and item descriptions iven. These descriptions indicated how individuals in a articular personality type were alike, and enabled the esearcher to name the types: the Self—Centered Person— lity and the Well—Integrated Personality. Results for ach of these processes are reported in Chapter IV. Multivariate Analysis of Variance The factor scores of a subject on the five learn— 9 environments were considered a vector response. Each 11 in the design contained vector responses of the dividuals classified in that cell. Two designs were ployed. L... 118 DESIGN ONE Personality Type Sex Self—Centered Well—Integrated Males Females L DESIGN TWO Teaching Levels Sex K—2 3—5 6—8 Special Ed; wales [I Females [ The design of this multivariate experiment paral— eled the familiar univariate case. In the statistical nalysis, the probability on the null hypothesis of the bserved mean difference between personality types, sex f intern teachers, and teaching levels for the five earning environments simultaneously was obtained by an {act multivariate test of significance. Univariate tests Juld have been performed on each variable separately. it a single probability statement applicable to all variables jointly could not generally be obtained from the separate tests. Because the five scores were obtained from the same subjects, they were correlated in some arbi— trary and unknown manner, and the separate F—tests would not be statistically independent. The multivariate test, on the other hand, was based on sample statistics which take into account the correlations between variables and have known exact sampling distributions from which the required probabilities could be obtained.39 In order to generate results which allow infer— ences to a specified population, the within—cell residuals must have the multivariate normal distribution with a common covariance matrix, and observations on different individuals must be uncorrelated. Unlike univariate analysis, multivariate statistical inference has not yet been proven robust to violations of these assumptions. Investigating the assumptions systematically through Monte Carlo methods is a formidable task even with 39R. Darrell Bock and Ernest A. Haggard, ”The Use of Multivariate Analysis of Variance in Behavioral Re— search,” Handbook of Measurement and Assessment in Behav— ioral Sciences, ed. by Dean Whitla (Reading, Mass.: Addison—Wesley Pub. Co., 1968), 102. 120 high-speed computers.40 However, the multivariate test is conservative. Conclusions of the investigation could be generalized with extreme caution, and only to popula— tions identical to the study sample. Scheffe' post hoc comparisons were made after the multivariate analysis of variance had shown significant differences in learning environments between male and female teachers and between teachers at different levels. Since twelve comparisons were made, a fiducial level of .01 was selected for the overall test and .001 for the separate tests.41 am A detailed description of the procedures of this investigation was presented in this chapter. Demographic 4OLyles Jones, "Some Illustrations of Psycholog— ical Experiments Designed for Multivariate Statistical Analysis,” A paper presented at a conference on Multi— variate Experimentation, Allerton Park (University of Illinois), November 15, 1960. 41 . . . Roger E. Kirk, Experimental DeSign: Procedures for the Behavioral Sciences (Belmont, California: Brooks/ Cole Publishing Company, 1968), 79—81. 121 data for the sample was given, instruments utilized in the study were previewed, and stages in the analysis were delineated. Finally, the statistical techniques employed in the study were described at length. Chap- ters IV and V provide the results of the analysis. W CHAPTER IV RESULTS OF THE FACTOR ANALYSIS Introduction The general nature of factor analysis and the two types, R and Q, used in this investigation were discussed in the previous chapter. Results of the exploratory factor analyses are presented here. Personality types are described first, then the learning environments are delineated. Personality Types Personality of a teacher is an important variable in the classroom, if not the most important variable. Numerous investigations on the subject of teacher per— sonality and teacher effectiveness indicate its interest to researchers. Teacher personality has been probed from many points of View: attitudes, values, interests, needs, beliefs, etc. Descriptions using one, a set, or even all 122 123 the constructs are still limited. Nevertheless, using a set of these constructs would provide a more valid picture of teacher personality than any single construct. Gordon hypothesized that the needs—system or pattern of needs of a teacher was more important than a single need when re— lated to classroom behaviors.l Parallel to her theory we could conjecture that the pattern of constructs, the organization of attitudes, needs, and beliefs contributing to a teacher's personality is more important than a single construct when related to overt classroom behaviors. Though each teacher has a unique personality, each teacher is a unique type, some teachers are more alike than others. Perhaps the organization of attitudes, needs, and beliefs within the personality is related to similarities observed. Types of teacher personality were explored in this study by analyzing score patterns on the Minnesota Teacher Attitude Inventory, the Edwards Personal Pneference Schedule and the Rokeach Dogmatism Scale. Eighty—eight lIra J. Gordon, "Assessment of Classroom Emotional Zlimate by Means of the Observation Schedule and Record,” Tournal of Teacher Education, XVII (Summer, 1966), 224— 226. 124 intern teachers were randomly selected from the total sample. The score patterns value for each subject was correlated with that of every other subject. This inter— correlation matrix was factor analyzed and a principal axis solution obtained. varimax rotations of this solu— tion produced two orthogonal factors accounting for 92% of the total common factor variance. Persons were clus- tered around two common syndromes of attitudes, needs, and beliefs; that is, each factor represented a person— ality type. Intern teachers were associated with one of the two personality types on the basis of their highest load- ing or correlation with a factor. Tables 4.1 and 4.2 summarize the subjects assigned to each type, their load— ing on each factor, and the communality (h2) or proportion of common factor variance explained in the rotated solu— tion. Sixty—two teachers correlated highest with the first factor (a) which accounted for 56% of the total variance. Fewer subjects (N = 26) loaded highest on the second factor (B) which explained an additional 36% of the total variance. 125 Table 4.1.——Subjects with Highest Loading on Factor d.* subject Loadings h2 Identification Factor a Factor 5 105 985 030 971 121 969 135 957 171 963 121 943 114 963 157 954 74 961 —033 925 108 960 184 955 75 957 000 916 104 954 "238 967 122 953 259 975 221 951 180 937 99 951 222 954 81 933 204 913 158 928 150 884 7 925 113 869 84 920 389 998 182 911 285 911 185 906 306 914 85 906 329 928 73 901 325 918 205 901 375 953 152 893 238 854 42 893 413 968 72 892 449 997 67 888 271 861 3 881 418 951 100 880 406 940 176 880 293 861 1 876 414 938 159 874 432 950 128 860 471 961 *Fifty—six percent of the total variance is accounted for by Factor a. 126 Table 4.1.——C0nt- subject Loadings h2 Identification Factor a Factor 3 175 836 519 968 193 836 501 950 160 835 "112 710 12 825 495 926 53 821 501 924 164 816 500 915 101 807 539 941 151 806 517 917 132 803 578 979 31 802 546 941 138 801 548 941 41 801 560 955 88 797 433 822 194 797 474 859 76 794 606 998 19 782 456 819 96 772 593 947 116 766 558 898 126 764 559 896 131 763 464 798 98 760 564 897 51 753 584 908 30 745 611 929 18 731 637 940 127 726 647 946 107 717 650 938 65 714 603 873 58 702 607 861 144 699 602 850 43 699 686 959 8 674 549 756 95 665 653 869 127 Table 4.2.——Subjects with Highest Loadings on Factor B.* Subject Loading hz Identification Factor a Factor 3 147 985 -081 977 22 979 —095 967 25 972 141 964 192 968 010 937 38 955 088 919 83 953 O6]- 912 35 946 022 896 186 941 140 905 201 938 299 969 120 930 103 876 106 922 262 920 17 920 346 966 157 877 476 996 134 874 398 922 174 864 426 928 169 831 436 880 48 813 462 875 20 786 587 963 199 783 469 834 102 781 600 970 92 775 545 897 213 760 538 868 214 739 630 942 94 714 619 893 32 695 632 883 154 694 655 910 *Thirty—six percent of the total variance is accounted for by Factor B. 128 Each personality type was described along sixteen dimensions: the fourteen manifest needs scores, the atti— tude and dogmatism scores. To do this, weighted standard scores for each of the sixteen variables for each subject were computed. The higher the factor loading of a sub— ject, the greater was the weight given to his raw score. These weighted values were summed across each variable separately and the arrays converted to standard scores. Tables 4.3 and 4.4 list the descending arrays of weighted scores for each variable on the first and the second per- sonality types respectively. Table 4.3.-—Descending Array of Standard Scores for Each of Sixteen Variables in the First Personality Type (Factor a). Variable Description Standard Score Order 0.73 Aggression 0.58 Endurance 0.54 Autonomy 0.53 Deference 0.51 Succorance 0.49 Dominance 0.36 Exhibition 0.34 Achievement 0.28 Heterosexuality 0.26 Nurturance 0.24 Affiliation 0.22 Change —0.01 Intraception —0.02 Dogmatism —1.89 NEAI -3.16 129 Type I: The Self— Centered Personality Persons loading highest on Factor d manifested a very high need for aggression, whether by attacking con— trary points of View, criticizing others, seeking revenge, or blaming others. These individuals had a high level of endurance, pushing themselves beyond the point of fruitful return. Though they needed to be quite autonomous they also could show deference to others: seeking suggestions, following instructions, giving praise, conforming to rules and regulations, accepting the leadership of others. In- directly, deference expressed in the right place and at the right time could provide them with even greater au— tonomy. Help, comfort, kindness, sympathy, and affection from others was a strong need. On the other hand, these persons also needed to provide such services to others, at least to a greater extent than individuals of the second personality type. Related to these needs for suc— corance and nurturance was an expressed need for affilia— tion. Friends were needed to give comfort, kindness, and encouragement; to maintain these friends, some return of affection, sympathy and kindness had to be shown. 130 Individuals in this type didn't express a strong need for ,self analysis or perception of others' needs (intraception was low). These subjects needed to dominate others, to be the center of attraction more than persons of the second personality type. But the most significant factors may have been their level of open-mindedness and their attitudes toward children and their profession. While belief—systems were more open than closed, the sixty—two individuals in this personality type tended to be more closed—minded than persons of the second type. Further— more, their attitudes Weighed heavily on the negative end of the favorable—unfavorable continuum. Self—centeredness 'seemed to be a common factor in the interrelation of these variables, hence this personality type was designated as Self—Centered. Table 4.4 presents the descending array of stand— ard scores for the sixteen personality variables in the second personality type. Interpretation of the array follows the table. 131 Table 4.4.—-Descending Array of Standard Scores for Each of Sixteen Personality Variables in the Second Personality Type (Factor B). Variable Description Standard Score MTAI 1.79 Heterosexuality 0.31 Achievement 0.28 Aggression 0.28 Deference 0.23 Dominance 0.22 Order 0.22 Autonomy 0.18 Exhibition 0.12 Succorance 0.08 Endurance 0.03 Change 0.02 Nurturance —0.02 Intraception —0.10 Affiliation —0.15 Dogmatism -3.51 132 Type II: The Well— Integrated Personality Individuals with highest loadings on Factor 8 expressed very positive attitudes toward children and teaching. Their belief—systems tended to be even more open than belief systems of the other sixty—two subjects. Heterosexual relations were slightly more important to the twenty-six persons of this type. They needed to achieve just as much as other individuals but expressed less need for autonomy, aggression, deference to others, or exhibition. These interns had a very low need for affiliation or attachment to friends. Consequently, praise, encouragement, sympathy, help, or understanding were not sought from others (low succorance); nor did they need personal fulfillment causing them to give af— fection or attention to others (low nurturance). They expressed less need for self—analysis or analysis of others' motives, feelings, and actions. They could let others direct their own destinies rather than predict their behaviors, categorize, or mold them. The common syndrome for this personality type seemed to be a well— integrated system of attitudes, needs, and beliefs. 133 Hence, this second personality type was denoted the Well- Integrated Personality. Table 4.5 includes the weighted standard score for each variable and the descending array of differences between personality types on a particular dimension. This array accents the distinguishing characteristics between the two types. Table 4.5.——Descending Array of Differences Between Standard Scores for Each of Sixtten PerSon— ality Variables in Factors d and 8. Standard Score Variable Description Difference Factor a Factor B Dogmatism —1.89 —3.51 1.62 Endurance 0.54 0.03 0.51 Order 0.73 0.22 0.51 Succorance 0.49 0.08 0.41 Affiliation 0.22 —0.15 0.37 Autonomy 0.53 0.18 0.35 Aggression 0.58 0.28 0.30 Deference 0.51 0.23 0.28 Nurturance 0.24 —0.02 0.26 Exhibition 0.34 0.12 0.22 Dominance 0.36 0.22 0.14 Intraception -0.02 —O.10 0.08 Achievement 0.28 0.28 0.00 Change —0.01 0.02 —0.03 Heterosexuality 0.26 0.31 —0.05 MTAI —3.16 1.79 -4.95 Subjects of the same personality type were‘ assigned a plus one or a minus one (dependent variable) for each type. Regression or least squares was used to assign normalized weights to each of the sixteen inde— pendent variables indicating their contribution in ac— counting for the variation of the dependent variable (personality type) above that accounted for by its mean. Thus, relationships between personality type and the sixteen personality variables were estimated. The re— gression coefficients obtained from the eighty—eight subjects were used to predict personality types for the entire sample. One hundred thirty—two intern teachers (74.16%) were classified in the Self—Centered Personality Type and forty—six individuals were predicted to have Well-Integrated Personalities. Classroom Learning Environments Different classrooms present unique situations because teachers, children, time, and events vary. Even the same classroom is ”new" each day because the people in it have had additional experiences. However, certain 135 events commonly occur in elementary school classrooms: teacher and pupils interact; materials are used; groups form and separate. Researchers of teaching behaviors have developed a variety of observational techniques to facilitate their search for effective patterns of class— room behavior.2 Individual behaviors which do not dif- ferentiate between teachers or classes are often combined into sets of behaviors or scales which do.3 I Factor analysis, which accomplishes parsimonious and meaningful description of data, was used to suggest kinds of learning environments observed in intern teachers classrooms. Total scores from six observations of each of 178 elementary intern teachers on one hundred variables of the OScAR III were intercorrelated. The resulting matrix was factor analyzed using a principal axis solution and varimax rotations. The fourth rotated solution was selected for further analysis after considering: 2 . . _. . . ' Simon and Boyer, Mirrors for BehaVior. 3Medley and Mitzel, "Measuring Classroom Behavior by Systematic Observation," 308—309. 136 1. The size of the eigenvalues; 2. The proportion of total variance eXplained by each factor; and, 3. The conceptual meaningfulness of the factors im— plied by the variables loading on each factor. Not all variables having significant loadings on a factor are reported in Tables 4.6 through 4.10. If the highest loading approached or exceeded .40, the variable was listed in the table; however, all variable loadings were used in computing the factor scores for each subject.4 In addition to the highest loadings, the second highest correlation, its related factor, and the communality for the variable are reported. Dimension I: Supportive Learning Environment The first dimension of classroom behaviors (re— ported in Table 4.6) manifested an atmosphere of mutual 4Variable loadings for the Fourth Rotated Solution and their communalities are listed in Appendix E. 137 support: teachers supported pupils, pupils supported the teacher and each other. The teacher engaged in a variety of activities: demonstrations and use of a variety of teaching aids. She was observed praising the good be- havior of children and also "talking to the class." This category included reproving remarks of the teacher re— garding pupil behaviors. Pupils demonstrated affection for both the teacher and other pupils, used a variety of learning aids rather than the textbook and were employed in art—related activities. 138 Table 4.6.-—Variables Loading Highest on Factor I (Sup— portive Learning Environment).* Highest Loadings Variable Number and Name Related 2 First Second Factor 15 Teacher demonstrates 694 —186 V 538 97 Subject: art, crafts 587 —234 IV 447 73 Teacher uses three dimensional object 579 154 IV 370 29 Pupil shows affection to teacher 576 163 IV 388 51 Pupil shows affection to pupil 553 —244 II 383 84 Pupil uses three dimensional object 536 —145 II 343 70 Teacher uses map, chart, picture 524 147 II 325 85 Pupil uses text, workbook -506 412 II 559 77 Teacher uses handi— craft, art materials 488 —186 IV 307 8 Teacher praises good behavior 479 -084 III 239 43 Pupil paints, cuts, draws 459 363 III 460 37 Pupil plays game 450 -166 V 240 88 Pupil uses handi— craft, art mate— rials 445 —340 IV 447 18 Teacher leads sing- ing, exercises, games 437 —238 V 299 14 Teacher illustrates with map, chart 407 226 II 230 12 Teacher talks to Class 393 236 III 267 *Seven percent of the variance is accounted for by Factor I. 139 Dimension II: Conventional Learning Environment Table 4.7 indicates the variables loading highest on Factor II. Traditional classroom activities were tapped in the second dimension. The teacher led the class by illustrating at the chalkboard and asking questions. Text, workbook, and chalkboard were the teaching aids utilized. Only two pupil activities loaded highest on this factor: reading aloud and using the chalkboard. This stereotype of a classroom was designated as a Con— ventional Learning Environment. Table 4.7.——Variab1es Loading Highest on Factor II (Con- ventional Learning Environment).* ' h t Loadin 3 Variable Number ng es 9 Related h2 Factor and Name . First Second 13 Teacher illustrates at board 634 —093 IV 414 33 Pupil reads aloud 621 —160 IV 457 69 Teacher uses chalkboard 613 —186 IV 433 79 No materials used by teacher —612 178 III 442 74 Teacher uses text, workbook 611 —448 I 600 1 Teacher questions, pupil answers 607 346 I 564 80 Pupil uses chalkboard 490 —185 IV 303 *Five percent of the total variance is accounted for by Factor II. 140 Dimension III: Independent Learning Environment Independence of pupils is strongly indicated by Factor III, the highest loadings of which are shown in Table 4.8. The teacher engaged in small group tasks while the majority of children pursued private study. Less than half the class worked with the teacher. Teacher assist— ance was given to individual pupils or their questions were answered. This dimension was named Independent Learning Environment. Table 4.8.—~Variab1es Loading Highest on Factor III (Independent Learning Environment).* Highest Loadings Related h2 Variable Number and Name -———————————————— Factor First Second 86 Pupil uses supplementary reading material 641 -120 IV 432 41 Pupil reads, studies at desk 613 -148 I 421 42 Pupil writes, manipulates at desk 591 234 II 469 87 Pupil uses writing material 584 273 II 429 60 Over one half the class in group task with teacher -568 099 I 344 9 Teacher works with indi- vidual pupil 561 325 I 456 2 Teacher answers pupil's question 540 453 IV 527 63 Over one half the class in ' d vidual task without in l 467 -227 IV 280 teacher *Five percent of the toal variance is accounted for by Factor III. 141 Dimension IV: Enriched Learning Environment Variables loading on Factor IV, reported in Table 4.9, appear to be unconventional and somewhat supportive if the second highest loadings are inspected. The teacher utilized some audiovisual aid during a science lesson or the pupil assumed the role of director of learning by dem- onstrating or using a special aid. Active learning on the. part of the pupil left little time for whispering that was not related to learning. Seven variables loading highest on this factor plus other significant correlations not in— dicated here characterized an Enriched Learning Environ— ment. Table 4.9.-—Variables Loading Highest on Factor IV (Enriched Learning Environment).* Highest Loadings Related h2 Variable Number and Name -—-—-——-—---- Factor First Second. 95 Subject: science 503 -073 II :2: 91 Subject: reading ~497 319 II 89 Pupil uses special learning aid 489 192 I. 338 34 Pupil demonstrates, illustrates 466 3:; i :8: 24 Pupil whispers -460 3 III 290 71 Teacher uses slides, film 460 ‘224 III 244 16 Teacher shows film, slide 431 -215 *Four percent of the total variance is accounted for by Factor IV. \_,.~ 142 Dimension V: Hostile Learning Environment A negatively charged emotional climate is de— scribed by the set of variables loading highest on the fifth factor. Pupils showed hostility to each other and to the teacher, interrupted the teacher or ignored ques— tions. On the other hand, the teacher also manifested hostility to pupils, ignored their questions, criticized or was sarcastic, warned or threatened the pupils. This dimension was denoted a Hostile Learning Environment. Table 4.10.——Variables Loading Highest on Factor V (Hostile Learning Environment).* Highest Loadings Related h2 Variable Number and Name —-—---—-—--- Factor First Second 26 Pupil interrupts teacher 667 145 II 498 50 Pupil shows hostility to pupil 577 173 I 395 28 Pupil shows hostility to teacher 557 ~238 I 380 4 Teacher ignores pupil's \ question 489 184 III 286 27 Pupil ignores teacher's question 487 -l32 II 287- 6 Teacher criticizes, uses sarcasm 485 -216 I 302 49 Pupil scuffles fights with pupil ' 448 ~29l II 311 3 Teacher interrupts pupil 442 100 II 213 22 Teacher leaves enters room I 404 -246 II 266 5 Teacher warns, threatens pupil 399 -236 I 285 *Four percent of the total variance is accounted for by Factor V. 143 Summary Factor analysis provided answers to two major questions posed for study. First, personality types were identified by investigating prospective teachers' patterns of response to paper—pencil inventories. The two types found in this research were designated the Self—Centered and the Well—Integrated Personalities. Second, distinct learning environments were observed in the elementary school classrooms of these teachers. Conceptual names given to the learning environments were: Supportive, Con— ventional, Independent, Enriched, and Hostile. CHAPTER V RESULTS OF MULTIVARIATE ANALYSIS OF VARIANCE Introduction Factor analysis provided a parsimonious descrip— tion of personality and classroom behavior variables: two personality types were identified from sixteen per— sonality scores and five learning environments were con— structed from one hundred categories of overt classroom behaviors. Results of these analyses posed a question regarding the relationships between certain independent variables such as personality type, sex, and teaching levels, and the five learning environments observed in an elementary school classroom. A five—variate multi— variate analysis of variance was performed using mean scores of (l) personality type by sex, and (2) sex by teaching levels. Significant findings were further in— vestigated using post hoc comparisons. The findings are reported in this chapter. 144 145 Personality Types and Sex Related to Five Learning Environments To determine the nature of differences (if any) in learning environments among male and female elementary intern teachers of two personality types, a five-variate multivariate analysis of variance was employed. Subjects were classified both by personality type and sex. Sub— class frequencies for the design are shown in Table 5.1. Table 5.l.——Subclass Frequencies of Male and Female Ele— mentary Intern Teachers in Two Personality Types. . Sex Personality Total Type Male Female Self—Centered 20 112 132 Well—Integrated 11 35 46 Total 31 147 178 The statistical hypotheses posed and tested were: Hypothesis One: There is no difference in learning en- vironments between teachers identified as having Self—Centered or Well— Integrated Personalities. Hypothesis Two: There is no difference in learning en— vironments between classes taught by male or female elementary intern teachers. 146 Hypothesis Three: There is no difference in -vhmnnem£ between classes learning en— taught by male or female elementary intern teachers identified as having Self—Centered or Well-Integrated Personalities. Results of the multivariate test are the next four tables. Mean product matrices Table 5.2; the overall tests of significance ity type, sex, and interaction are exhibited presented in are shown in for personal— in Table 5.3. Table 5.2.-—Mean Products: Multivariate Analysis of Variance of Learning Environments for Personality Types by Sex. MT: Between personality types df = l 8 1.182 " C -0.374 0.118 I -l.888 0.598 3.015 E -0.678 0.215 1.083 0.389 H -0.414 0.131 0.661 0.237- 0.145 MS: Between sexes df = 1 s _ 1.602 C 3.483 7.568 I -l.039 —2.258 0.674 E 3.444 7.485 -2.234 7.404 H -l.4l7 —3.080 0.919 -3.046 1.253 ' ' df = l MTS' Interaction - s f 0.000 C —0.005 0.142 I -0.002 0.040 0.011 E 0.020 —0.511 -0.145 1.844 H 0.009 —0.230 —0.065 0.829 0.372 Me: Within-cell df = 174 s ' 0.999 C -0.259 0.986 I 0.020 0.196 .1.009 E -0.268 0.188 0.104 0.954 H -0.290 0.183 0.079 0.292 1.013_ S: Supportive C: Conventional I: Independent E: Enriched H: Hostile j 147 Matrices of personality type and interaction (Table 5.2) show a structure similar to that of the re— sidual covariance matrix (error term). Neither person— ality type nor interaction of personality type and sex was significant, as the overall test results indicate (Table 5.3). Table 5.3.-~Tests of Significance in Multivariate Analysis of Vari= ance for Two Personality Types of Male and Female Ele-= mentary Intern Teacherso Effect p = 5* Type Sex Interaction nh# l l 1 net 174 174 174 F 0.826 5.119 0.508 df 5 and 170 5 and 170 5, 170 prob. p < 0.5327 p < 0.0003 p < 0.7701 Discriminant function, sex: V = -. ~ . + . - . ’ . S 0 498S 0 674C 0 325I 0 715E + O 376H *p equals the number of dependent variables: Supportive, Conven~ tional, Independent, Enriched, Hostile. #degrees of freedom associated with Mh, or the matrix of mean products for a classification variable or an interaction. tdegrees of freedom associated with Me, or a p x p matrix of error mean products. 148 Results of the multivariate test indicate that only one hypothesis was supported at the chosen level of significance (.05). Hypothesis One: There is no difference in learning environments between teachers iden— tified as having Self— Centered or Well— Integrated Personalities. Fail to reject. Hypothesis Two: There is no difference in learning environments between classes taught by male or female ele— mentary intern teachers. Reject (p < .01) Hypothesis Three: There is no difference in learning environments between classes taught by male or female elementary intern teachers identi— fied as having Self— Centered or Well— Integrated Personalities. Fail to reject. No significant differences were found between teachers classified as Self-Centered or Well—Integrated in their style of teaching. The teaching environments of males were significantly different from those of females. Male teachers were observed in more Supportive 149 (.234 versus —.O37), Conventional (.442 versus —.092), and Enriched (.416 versus —.107) learning environments, while females were seen in more Independent and Hostile class— room settings. When the interaction of sex and person— ality type (Hypothesis Three) was analyzed, no significant differences were observed beyond those attributable to sex. Means for these variables are presented in Table 5.4. Figure 5.1 graphically describes these differences. Figure 5.1 illustrates, too, that females tended toward less variability in learning environments observed in their classrooms. Table 5.4.-—Subc1ass Means of Male and Female Elementary Intern_ Teachers in Two Personality Types. Sex Learning Personality Type Means EnVironment . Self—Centered Well-Integrated Supportive 0.174 0.343 0.234 Conventional 0.524 0.292- 0.442’ Male Independent -0.047 -0.366 -0.l60 Enriched' 0.310 0.608 0.416 Hostile -O.254 -0.096 -0.l98 Supportive -0.075 0.087 -0.037 Conventional -0.075 ~0.148 -0.092 Female Independent 0.101 -0.l73 0.036 Enriched -0.042 -0.316 —O.107 Hostile 0.053 -0.047 0.030 Supportive -0.037 0.148 0.011 Conventional 0.016 —0.043 0.001 Means Independent 0.079 -O.219 0.002 Enriched 0.083 -0.095 0.037 Hostile 0.006 -0.059 0.011 Lu 150 4, -———Supportive - ..».Conventional .3 ———-Independent 2 —~—Enriched “ -~»-Hostile .1 O —.l —.2 _.3 -.4 Male) Female Figure 5.l.——Comparison of Overall Means for Male and Female Elementary Intern Teachers on Each of the Learning Environments. To further investigate the source of differences, the linear function of the dependent variates which maxi— mally discriminates between sexes was computed. The dis— criminant function, for which coefficients are provided by the latent vector associated with the significant latent root for sex effect, was as follows: v =—. —. .2 —.' . . 0498S 0674€+o35I O715E+0376H Learning environments contributing the most to the func— tion had negative weights: Enriched, Conventional, and 151 Supportive. Mean scores for males were higher on each of these dimensions. To support the previously computed analyses the step—down F test was employed. The results of this test are exhibited in Table 5.5. In this test, the investi— gator usually orders dependent variables and critical values according to his priority of interest. Order of the dependent variables was established by the successive extraction of each factor in the factor analysis. Almost equal amounts of variance were accounted for by each factor, thus justifying use of equal a levels for each dependent variable. The sum of the a levels for the de— pendent variables equals the chosen fiducial level for the overall test of significance (.05). Conventional and Enriched learning environments (the same dependent variables contributing the most weight to the discrimi— nant function) were significant step—down F statistics, and thus accounted for the major differences between male and female intern teachers. 152 .mm.o we Eoeomnm mo momnmme oNH\H How 05Hm> HMUflpHMU .Hm>mH Ho. egommp ucmonencmnme so. oee\s ee.m esH\H em.H wsnpmom .Ho. HeH\H NH.m esH\H we.e ewsonncm so. meH\H em.s ees\s no.0 unmecoamecH .Ho. msH\H os.os ees\s mw.e Hchnnam>coo Ho. eeH\H oo.s eeH\H oo.H m>nunocmsm o Eocmmum Mo m EOUmwum mo m 2% mucmficoufl>cm mmwummm CBOUImmum mmmumwm mumflum>flcb mcflsuqu .uommmm xom mo umme czoenmmumnn.m.m magma h 153 Teaching Level Related to Five Learning Environments The same five—variate multivariate analysis of variance was utilized to determine the nature of differ— ences (if any) in learning environments among male and female elementary intern teachers at four teaching levels: K—2, 3—5, 6—8, and Special Education. Subclass fre— quencies for the 2 x 4 design are provided in Table 5.6. Table 5.6.——Subclass Frequencies of Male and Female Ele— mentary Intern Teachers at Four Teaching Levels. Teaching Levels Sex Total K-2 3-5 6-8 Spec. Ed. Male 1 l3 l4 3 31 Female 71 56 13 7 147 Total 72 69 27 10 178 Two statistical hypotheses were posed and tested: Hypothesis Four: There is no difference in learning en— vironments among classes in teaching levels K—2, 3—5, 6—8, and Special Education. he 154 Hypothesis Five: There is no difference in learning en— vironments between classes taught by male and female elementary intern teachers at teaching levels K—2, 3-5, 6—8, and Special Education. Results of the multivariate analysis of variance for mean vectors (shown in Tables 5.7 and 5.8) indicate that both main effects were significant at greater than the .01 level of confidence. The mean products (Table 5.7) between the four teaching levels on each dependent variable show a pattern of differences counter to the structure of the within—cells covariance matrix (Me). Differences lay in both the size and sign of variance and covariance coefficients. Iliilw 155 Table 5.7.—-Mean Products: Multivariate Analysis of Var— iance of Learning Environments for.Sevay Teaching Level. ‘7 MS: Between sexes df = l S 1.875 C 3.700 7.301 I —1.395 —2:618 0.985 E 3.621 7.144 -2.624 6.991 H —l.578 —3.113 1.143 —3.046 1.328 ML: Between teaching levels df = 3 S C 0.429 I 0.258 0.687 E 1.459 —0.584 9.601 H 0.679 1.089 1.060 2.575 : ' df = 3 MSL Interaction S 0.540 C 0.073 0 314 I —0.907 0 317 2.359 E 0.038 0.237 0.371 0.229 H -0.289 —0.538 -0.343 -0.453 1.029 Me: Within-cells df = 170 S 0.781 C —0.243 0.999 I 0.036 0.197 0.995 E —0.262 0.163 0.118 0.819 H —O.192 0.184 0.073 0.485 0.976 S: Supportive C: Conventional I: Independent E: Enriched H: Hostile Table 5.8.—-Tests of Significance in Multivariate Analysis of Variance for Male and Female Elementary Intern Teachers at Four Levels of Teaching. Effect p* = 5 Sex Level' Interaction nh# l 3 3 neT 170 170 170 F 6.686 6.718 1.130 df 5, 166 15, 459 15, 459 prob. p < .0001 p < .0001 p < .3266 Discriminant functions = _ _ , . — 0.85 + 0.4 6 VseX 0.704S 0 649C + O 397I 5E l H V _ t1l = — 1.081S — 0.227C + 0.310I 0.765E + 0.300H V t12 = 0.414S + 0.061C + 0.036I - 0.945E + 0.058H *p = the number of dependent variables: Supportive (8), Conventional (C), Independent (I), Enriched (E), and Hostile (H). #nh is the degrees of freedom associated with Mh' the matrix of mean products for a classification variable or an interaction. Tne is the degrees of freedom associated with Me, the p x p matrix of error mean products. 157 Results of the hypothesis testing showed support for one hypothesis. Hypothesis Four: There is no difference in learning environments among classes in teaching levels K-2, 3—5, 6—8, and Special Education. Reject (p < .01) Hypothesis Five: There is no difference in learning environments be— tween classes taught by male and female elemen— tary intern teachers at teaching levels K—2, 3—5, 6—8, and Special Educa- tion. Fail to reject. Since the interaction of sex and teaching level failed to reach significance and the sex main effect has been pre— viously analyzed and discussed, teaching level differences will be examined in detail. Subclass means for the sex by teaching level de— sign (Table 5.9) were used to construct the graph for mean scores of each teaching level on each dependent variable (Figure 5.2). Notable differences were found in the Sup— portive and Enriched dimensions of classroom behaViors. Supportive learning environments were observed more often 158 Table 5.9.——Subclass Means of Male and Female Elementary Intern Teachers at Four Teaching Levels. Level Learning Male Female Means EnVironment Supportive —l.122 -O.232 ~0.244 Conventional —0.09l —0.091 ~0.09l K—2 Independent 1.400 0.104 0.122 Enriched —0.480 —0.434 -0.435 Hostile 0.778 0.068 0.078 Supportive 0.192 0.275 0.259 Conventional 0.492 ~0.l78 —0.052 3-5 Independent 0.143 —0.l42 —0.088 Enriched 0.184 —0.019 0.019 Hostile —o.4oo —0.008 —0.082 Supportive 0.773 0.453 0.619 Conventional 0.432 0.038 0.242 6-8 Independent —0.528 0.222 —0.l67 Enriched 0.601 0.562 0.582 Hostile ~0.388 —0.258 —0.325 Supportive -1.650 —l.457 —l.515 Conventional 0.446 0.336 0.369 Sp. Ed. Independent —0.279 0.416 0.208 Enriched ‘0.851 1.261 1.138 Hostile 1.236 0.476 0.704 Supportive 0.234 —0.037 0.011 Conventional 0.442 —0.092 0.001 Means Independent —0.l60 0.036 0.002 Enriched 0.416 —0.107 0.037 Hostile —0.l98 0.030 0.011 159 1_2 --—- Supportive 1.0 ....... Conventional "l“, Independent “I. .8 -u_n Enriched u’ .6 _x_" Hostile /$,“/ // .4 I .2 0 —.2 —.4 —.6 -.8 —1.0 _1.2 —1.4 -l.6 K—2 3—5 6—8 Spec. Ed. Figure 5.2.——Comparison of Overall Means for Teaching Levels of Elementary Intern Teachers on Each of the Dependent Variables. as grade level increased in regular K—8 schools. Special Education classes had low scores on the Supportive behav— iors dimension. Probably the dependence of Special Edu— cation children on their intern teacher explains some of this low score. Mutual support between teacher and pupils or among pupils themselves may be related to the age and 160 maturity of the individuals involved. On the other hand, Enriched learning environments were observed most fre~ quently in Special Education classes and declined in fre— quency from grades 6-8 down through K—2 teaching levels. Extensive use of multi—media in Special Education is no surprise, but the fact that primary teachers were less often observed in Enriched environment is notable. Per— haps many observations were made during reading classes in the primary levels, a fact which could account for the differences. Each significant discriminant function for teach— ing levels was used to calculate discriminant scores for all subjects. Two separate sets of scores were obtained, the first set from the first function, the second set from V . t12 thl = —l.081s — 0.227C + 0.310I — 0.765E + 0.300H thz = 0.414S + 0.061C + 0.036I — 0.945E + 0.058H According to the first function, Supportive and Enriched versus Independent and Hostile learning environments ac— counted for most of the difference in behaviors across 161 teaching levels (see Table 5.11,‘step—down test of teach— ing level effect). On the second function, Enriched en— vironments carried the most weight (negative) while all other weights were positive. Mean discrimination scores (Table 5.10) for each teaching level indiCate the distinct differences between these functions. Table 5.10.——Mean Discriminant Scores for Four Teaching Levels. Discriminant Teaching Levels score ' K—2 3—5 6—8 Spec. Ed. First 0.678 —0.335 —l.319 0.959 Second 0.313 0.078 -0.304 —l.631 The first discriminant score for teaching levels con— trasted primary and Special education classes with later elementary grades (3—8). Special education and upper elementary grades were contrasted with primary and middle elementary classes in the second set of discriminant scores. Twelve post hoc comparisons between mean discrim— inant scores for selected teaching levels were made on o. eeommn unmoflmncmnme . .Hm>ma H illn||11|ll||||r1nl\anlutxl|x|||\:\\|‘l|\11 Ho. ems\m Ne.s oss\m we.m manumom *Ho. sms\m mo.eH oes\m me.HH emsonncm m so. mma\m os.s ose\m 86.0 nemecmmmecH so. mos\m He.o oee\m me.o HmcoHucm>coo .Ho. oeexm Hm.ss oss\m Hm.eu m>nunomd5m o EOUome mo m . EOUmem mo m , mbcmficoufl>cm monsoon CBOUImmum mmmnmmm wumflum>flcb mcflcummq .uowwmm H®>mfl mcfl£ommB mo #mOE CBOUIQmumII.HH.m THQMB 163 each set. A fiducial level of .001 was chosen for sig- . 1 . . . . nificance. The critical value was determined according 2 to the formula: F'-'=(k—1)F. a/VlIVZ where: F = the tabled value for d = .001 k = 4 treatment levels vl=k-l=3 v2 = N — k = 178 — 4 = 174. 120 degrees of freedom were used. Thus: F' = 3 (11.38) = 34.14 Tables 5.12 and 5.13 summarize the results of twelve post hoc comparisons of teaching levels on the first and second discriminant scores respectively. lRoger E. Kirk, Experimental Design: Procedures for the Behavioral Sciences (Belmont, California: Brodks/ Cole Publishing Company, 1968), 79—81. According to the Bonferroni inequality, the overall level of a, say aALL'~ is equal to kd when k tests have been done at the d level. In this case, k = 12 and Q’ALL 74.012. 2Ibid., 90—91. 164 Table 5.12.-—Post-Hoc Comparisons of Means for Teaching Levels Using the First Discriminant Score. Source of Variance df Between F—ratio Mean:5quare Grades K—8 Special Education 1 9.77 9.59 Grades K—2 : Grades 3—8 1 51.04 50.07* Grades K—2 : Grades 3—5 1 36.21 35.51* Grades K—2 : Grades 6—8 1 77.57 76.09* Grades 3—5 : Grades 6-8 1 18.76 18.40 Grades K—2 : SpecialTEducation 1 .69 .68 Grades 3—5 Special Education 1 22.98 22.54 Grades 6—8 Special Education 1 8.58 8.41 Grades K—2 Spec. Ed. : Grades 3—8 1 65.47 64.22* Grades K—5 : Grades 6—8 1 25.65 25.16 Grades K—2 Spec. Ed. : Grades 6-8 1 40.38 39.61* Grades K—2 Spec. Ed. : Grades 3—5 1 30.99 30.99 *Significant beyond the .001 level. Contrasts between primary and middle—upper elementary levels, between primary—special education classes and middle—upper elementary grOups were significant beyond the .001 level. None of the comparisons on the second set of discriminant scores was significant (Table 5.13). L. 165 Table 5.13.——Post Hoc Comparisons of Means for Teaching Levels Using the Second Discriminant Score. Source of Variance df Between F—ratio Mean Square Grades K—8 : Special Education 1 28.85 29.40 Grades K-2 : Grades 3—8 1 5.69 5.80 Grades K—2 : Grades 3-5 1 1.95 1.98 Grades K—2 : Grades 6—8 1 .49 .50 Grades 3—5 : Grades 6—8 1 2.82 2.87 Grades K—2 : Special Education 1 33.18 33.81 Grades 3—5 : Special Education 1 21.68 22.09 Grades 6—8 : Special Education 1 7.59 7.74 Grades 3-8 : Grades K—2, Special Education 1 7.22 7.35 Grades K—5 : Grades 6—8 1 23.47 23.92 Grades 6—8 : Grades K-2, Special Education 1 .38 .38 Grades 3—5 : Grades K—2, Special Education 1 12.64 12.88 S_um_me£r Multivariate analysis of variance was employed to test the significance of differences in learning environ— ments for personality types, sex, and teaching levels. Sex and teaching levels were both significant at greater than the .01 level of confidence. Male teachers were observed in relatively more Conventional, Enriched, and Supportive learning environments. Female intern teachers i had negative overall mean scores on these dimensions and slightly positive overall means for Independent and Hos— tile learning environments. Special education classes were obviously different from classes in the K—8 organi— zational system. Learning environments for special edu— cation interns were more Enriched and less Supportive than at any other level. Grades 6—8 portrayed the most Supportive learning environments and primary classes the least Enriched. CHAPTER VI CONCLUSION OF THE INVESTIGATION Introduction Four major topics are discussed in this final chapter. After summarizing the investigation, limiting factors are presented. With these limitations in view, conclusions are derived from the study. Implications of the investigation are considered in the concluding section of the chapter. Summary of the Investigation The researcher investigated the relationship be— tween two forms of observation of teacher behavior: (1) the teacher's perceptions of his own attitudes, needs, and manner of believing, and (2) an external source's perceptions of overt behaviors in the teacher's elementary school classroom. It was conjectured that accurate re— ports of self—perceptions (indirect observations) and 167 168 unbiased records of overt classroom behaviors (direct observations) would be congruent. That is, personalities of teachers would be consistent with the learning environ— ments observed in their elementary school classrooms. Teachers expressing positive attitudes toward children, open belief systems, less need for order, less dominance or aggression were expected to be more flexible and sup— portive in their classrooms than teachers having opposite or near Opposite characteristics. Data were gathered in two phases from 178 parti— cipants in the Elementary Intern Program of Michigan State University. Subjects entering EIP in Fall, 1967 and Winter, 1968, reSponded to three personality inventories: the Minnesota Teacher Attitude Inventory, the Edwards Personal Preference Schedule, and the twenty—item Rokeach Dogmatism Scale. Thirty—seven trained observers recorded overt classroom behaviors of these individuals during the final six weeks (May—June, 1969) of their teaching intern— ship in public elementary schools of southern Michigan. The Observation Schedule and Record (OScAR III) was em— ployed as a standardized recording form for each of the six observations per intern teacher. Factor analytic techniques were applied to parsi— moniously describe the sixteen personality scores (Q—type analysis) and one hundred categories of OScAR III (R—type analysis). Two personality types were identified by analyzing the factor loadings of individuals on the first rotated solution which accounted for ninety—two percent of the total variance. The personality types were denoted Self—Centered (132 interns) and Well—Integrated (46 sub— jects). The total scores on overt classroom behaviors for the 178 intern teachers were intercorrelated, a factor matrix formed, and the principal axis solution rotated orthogonally. By analyzing the factor loadings of each factor in the fourth rotated solution (selected as more parsimonious yet accounting for at least one—fourth of the total variance) five learning environments were de— scribed: Supportive, Conventional, Independent, Enriched, and Hostile. Factor scores were computed for each sub— ject on each of the five factors. A five—variate (one for each learning environment) multivariate analysis of variance was employed to test the significance of differ— ences in mean vectors for each of the following hypoth— eses 2 Hypothesis One: Hypothesis Two: Hypothesis Three: Hypothesis Four: Hypothesis Five: 170 There is no difference in learning environments between teachers iden— tified as having Self— Centered or Well— Integrated personal— ities. Fail to reject. There is no difference in learning environments between classes taught by male or female ele— mentary intern teachers. Reject (p<.Ol) There is no difference in learning environments between classes taught by male or female ele— mentary intern teachers identified as having Self—Centered or Well— Integrated personalities. Fail to reject. There is no difference in learning environments among classes in teach— ing levels K—2, 3—5, 6-8, and Special Education. Reject (p<.Ol) There is no difference in learning environments betWeen classes taught by male and female ele— mentary intern teachers at teaching levels K—2, 3—5, 6—8, and Special Education. Fail to reject. 171 Discriminant score analysis indicated that male intern teachers were observed more often in Conventional, Supportive, and Enriched and less frequently in Indepen— dent and Hostile learning environments than were female intern teachers. Scheffe' post hoc comparisons of mean discriminant scores resulted in significant differences (p < .001) between the following teaching levels: 1. Grades K—2 portrayed more Independent and Hostile, less Enriched, Supportive, and Conventional learn— ing environments than Grades 3—5, Grades 6—8, and combined intermediate and upper elementary teach— ing levels. 2. Early elementary and Special Education classes Were observed with much less Supportive, more Enriched, Hostile, and Independent learning en— vironments than intermediate and upper elementary classes combined or upper elementary grades alone. l72 Limitations of the Investiqation This research was conducted in a field setting rather than the scientific educational laboratory. There— fore many limiting factors must be enumerated before con— clusions derived from the study can be stated, or general— izations drawn from them. Data Collection A major problem in this study was the time span over which data were collected. Two distinct time periods were involved. During Fall Term, 1967 and Winter Term, 1968, subjects responded to items on the three personality inventories. What has been termed indirect observations, that is, self—perceptions of the pre—intern regarding his attitudes toward children and teaching, his manifest needs and beliefs Were assessed. Direct observations (percep— tions by an outside source) were recorded during a six— week period from May—June, 1969. T1 = time of indirect observations = Fall, 1967 to Winter, 1968; T2 = time of direct observations = May—June, 1969. The time span covered eighteen months for subjects entering EIP in Fall, 173 1967 and fifteen months for students taking their methods courses in Winter, 1968. Direct Observation of Overt Classroom Behaviors Probably no classroom observation study follows all the guidelines established for such research. Attempts must be made to approach the ideaL however, and failure to do so must be reported. Selection of an observation form appropriate to the nature of the problem is a critical step in a class- room observation project. Several advantages weighed heavily in choosing the Observation Schedule and Record (OScAR III and 5V). First, the instrument was originally designed to gather data on the classroom behaviors of beginning elementary school teachers. The subjects of this investigation were teaching full time as beginning teachers with supervision, and thus appeared to be com— parable. Second, categories were of such a nature as to be evidently present or not, thus observer training was facilitated. Third, the process of observing was sep- arated from the process of scoring. No judgment of 174 appropriateness of behaviors was required of the observers. They merely checked all behaviors which occurred in a five—minute interval. Verbal interaction between the teacher and pupils was not tapped by the OSCAR form em— ployed in this study. Thus the extent to which verbal behaviors affect classroom learning climate would limit the generalizability of the results. Another limitation regarding the classroom obser— vations lay in the training procedures. Aside from the short period (four hours» filmed episodes rather than actual classroom settings were utilized in learning to re— cord the overt behaviors. The number of observers being trained hindered practice in real elementary school class— rooms. The films used in training the observers approxi— mated actual classrooms since the filmed episodes were developed specifically for such work. Methods of observing employed in this study may have introduced other biases. The number of observers (N = 37), the fact that each intern consultant observed the intern teachers assigned to him from the beginning of the year, and that observers were not crossed with 175 subjects or the ten centers all could have contributed unknown biases into the investigation. Other possible intervening variables can be nu- merated: location of the EIP centers, the school set— tings, the socio—economic status of both intern teachers and their students. These elements, when uncontrolled as in this study, may easily introduce biasfinto.therre— search. Statistigal Techniques Analysis of variance is generally considered a very strong statistic to employ when analyzing data. Findings are easily generalized because the test is robust to violations of the assumptions of normality and homo— scedasticity. But multivariate analysis of variance is not robust to violations of these assumptions. Very com— plex tests are required to prove the assumptions true for the data being analyzed. Since these tests were not performed, results of this investigation must be general- ized with care to comparable groups of subjects. 176 With these cautions in mind, then, the conclusions of this study will be stated. The reader should keep them in view when considering conclusions and their implica— tions. Conclusions Conclusions drawn from the study are listed below. Factor analytic techniques were employed to parsimoniously and meaningfully describe person— ality and classroom observation variables. Individuals expressing positive attitudes toward children and teaching, Open—mindedness in their belief systems, less need for ag— gression, dominance, exhibition, order, en— durance, succorance, affiliation, autonomy, deference, nurturance, and intraception were distinguishable from teachers not having these characteristics. 177 b. Distinct learning environments were identified by analyzing total scores of intern teachers on OScAR III observation records of overt be- haviors in their elementary School classrooms. Learning environments of Well—Integrated intern teachers did not differ significantly from learn— ing environments of Self—Centered intern teachers identified in this study. Learning environments of male intern teachers of this sample differed significantly from those of female intern teachers. Elementary school class— rooms of male teachers.were observed as more Sup— portive, Conventional, and Enriched while female intern teachers were observed in more Independent and Hostile learning environments. Learning environments were significantly different in various teaching levels for this sample of in- tern teachers. a. Grades 6—8, 3-5, and 3~8 were observed to have more Enriched, Supportive, and 178 Conventional learning environments than K-2 classes. b. Grades K—2 and Special Education classes com— bined were observed as more Enriched, Hostile, and Independent learning environments than grades 6—8 or grades 3«8. The latter teaching levels shOWed more Supportive learning envir— onments. Though more Independent learning environments were ob— served among female intern teachers and in primary (K—2) classes where the ratio of female to male teachers was 71/l; and though more Supportive learning environments were observed among male teachers and in upper elementary (6—8) classes where male teachers outnumbered female teachers, no significant interaction between sex and teaching level was obtained. Implications of the Investiqation Teaching personnel problems demand as much atten— tion today as they have in the past. Large numbers of 179 teachers leave the profession for various reasons. Still, little is known about the comparative effectiveness of those who remain and those who leave. The surplus of teachers on the job market allows teacher education in— stitutions to be more selective among applicants. Yet teaching potentialiis difficult to determine without fur— ther knowledge of the factors which constitute successful teaching. Rising cost of living and collective bargaining by teacher organizations have brought increases in teacher salaries. Laymen demand that increased expenditures for education bring increased returns in teaching effective— ness. Salary schedules and promotions of personnel are usually based on some estimation of teaching merit. Fed— eral monies have been expended to stimulate development of comprehensive model teacher education programs.1 Follow—up studies of teaching effectiveness are needed to evaluate these changes in college curriculums. Each of these factors requires assumptions about what constitutes good teaching and predictions regarding lJoel L. Burdin and Kaliopee Lanzillotti, A Reader's Guide to the Comprehensive Models for Preparing Elementary Teachers (Washington, D.C.: Eric Clearing— house on Teacher Education, 1969). 180 alternate courses of action in achieving effective teach— ing. Often assumptions and predictions are not directly, much less explicitly, stated. Furthermore, their accuracy is not always checked. This investigation suggests pos— sible alternatives for study of teacher effectiveness in teacher education, selection and placement of personnel, supervision and in—service education of teachers, and further research on teaching. Teacher Education Significant differences in learning environments between male and female elementary intern teachers and between various teaching levels in the elementary school suggest the need for differential preparation of teachers. Programs should be personalized to meet the needs of stu- dents according to their sex and the grade level they plan to teach. Internship in various positions on the school staff may help future teachers select jobs for which they feel most competent. Teachers, university personnel, and members of the community should design teacher education programs 181 which develop competencies future teachers will need. Interns who can create and maintain supportive and en— riched learning environments and promote independent learning as pupils mature will certainly meet the chal— lenge of educating American youth. No personality pattern has been demonstrated as essential for teaching, though we might suspect a Well— Integrated Personality to be more effective than a Self- Centered Personality. Teacher educators should create an atmosphere wherein students can assess their own per— sonality and attitudes and find guidance in modifying components they wish to improve. The usefulness of observing specific behavior cues as a criteria for teacher effectiveness has been demonstrated. Teacher education candidates should be exposed to many observation schedules, become efficient in the use of one or more as a means of self—appraisal, and finally design their own system consistent with their philosophy of teaching and education. 182 Recruitment and Placement Male intern teachers exhibited more Supportive, Enriched, and Conventional classroom learning environments than did female elementary intern teachers. If Supportive and Enriched learning environments are desirable in ele— mentary school classrooms, two suggestions might be made. First, more male teachers should be recruited for elemen— tary school teaching positions. Second, elementary school female teachers should be guided in creating and maintain— ing Supportive and Enriched learning environments. Female intern teachers were observed in more In- dependent and Hostile learning environments. Most female interns of this sample taught in primary and intermediate levels where the teacher is frequently engaged with smaller groups of pupils while others pursue independent activities. Are primary children able to work indepen— dently or should such activities be less frequent in early elementary grades and increase as the child progresses through school? Perhaps class size should be reduced on the primary level or several teacher aides per class should be employed to guide the independent work. 183 Tpacher Supervision and In—Service Education Intern consultants assist five or six intern teachers (participants in the Elementary Intern Program) throughout their first year of teaching. If this rela— tionship is perceived as a "helping" rather than coer— cive interchange, beneficial results should accrue. Con— sultants should be aware of the attitudes and personal needs of their interns, the assets and limits imposed by such attitudes and personal qualities. Differences in learning environments at varidus levels should be under— stood by consultants. Nor should they expect male and female interns to teach the same way. They should guide intern teachers in techniques of self-appraisal of their attitudes toward children and teaching, their needs and beliefs, and also overt teaching behaviors and their re- lationship to personal qualities. Consultants could guide female intern teachers in developing more Supportive and Enriched learning en— vironments. Hostility would probably decrease if emphasis were placed on the positive approach. 184 Intern consultants would probably enhance their effectiveness if specific behaviors were observed regu- larly. An observation form like the OScAR could provide objectivity to their visits. During the internship year ten visits (one per month) could be devoted to use of a specific form agreeable to both intern and consultant. A follow—up conference comparing observations could be held, encouragement given, and suggestions for improve— ment proposed by the intern. Suggestions for Further Research The theory upon which this research was based, that a congruent relationship exists between direct and indirect observations of teacher behavior, was not sup— ported in this investigation. The major problem, however, may have been the data collection procedures. It is probably unreasonable to assume that an individual's self—perceptions at T1 would be identical to his self— perceptions at T (1—1/2 years later) when many eXper— 2 iences have intervened. Research should be undertaken to test the congruent relationship between teachers’ 185 self—perceptions and observations of their overt teaching behaviors when assessments are made simultaneously. Three personality types might be identified in future studies: Self—Centered, Well—Integrated, and Mixed. The criteria used in assigning persons to types could be (1) the highest factor loading exceeds .40 while the second loading is less than .40; or (2) the square of the highest factor loading approaches the communality. The latter criterion requires almost all of the common factor variance to be explained by the factor to which the individual is assigned. If neither of these criteria can be met, subjects could be assigned to the Mixed Type. Significant differences in overt teaching behaviors might have been observed if individuals loading almost equally on the two factors had been placed in a third category. . Numerous limitations noted previously suggest other possibilities for classroom observation research. More control of observers and methods of observing would be gained by drawing a random sample of subjects, reducing the number of observers and increasing the amount of ob— servation time. Each observer could record behaviors of each teacher included in the sample, perhaps by using 1 “A 1 1 l 186 video tapes of classroom behaViors. Records of verbal behavior should be included in the observations to pro- vide more complete records of classroom behaviors. This study demonstrated that teaching is a multi— dimensional activity. The five factors identified ac— counted for only 25% of the total variance; hence much of the teaching—learning situation was not described. Research is needed to determine other significant compo— nents of teaching. Two interesting relationships were observed in the Supportive and Hostile learning environments across the four teaching levels. Supportive classroom settings were observed least often in Special Education and early elementary classes were seen more often in Grades 3—5, and most of all in upper elementary levels. However, the reverse situation was true of Hostile learning en— vironments. Frequency of occurrence decreased as grade level increased. Is this the result of the socialization process? Is more control necessary in Special Education and primary classes in order to condition the pupils to "appropriate social" behaviors? As overt hostility de- creases, does covert hostility increase? Does hostility 187 reach its peak among college students though in a sup— pressed form? Personality type, sex, and teaching level were the only independent variables investigated in this pro— ject. Age and marital status of the teacher, socio— economic status of both teacher and pupils, location of the school (inner city, urban, suburban, or rural) could also be related to learning environments observed in ele— mentary school classrooms. Further research might well include any or all of these additional variables. E Elementary school teaching seems to be in transi— g tion. Learning environments in "classrooms of the future"_ may differ radically from the five types identified in this project. Observation recording forms which appear appropriate today may lose validity and reliability simply because the behaviors included seldom occur. Five—year and ten—year longitudinal studies of a random sample of teachers included in this investigation could provide data on the nature of the change in teaching—learning environments of elementary schools and the changes that occur in teacher—behaviors over time. SELECT ED B IBL IOGRA PHY SELECTED BIBLIOGRAPHY Books and Pamphlets Allen, Robert M. Variables in Personality Theory and Personality Testing. Springfield, Illinois: Charles C. Thomas Publisher, 1965. The Association for Student Teaching. Internships in Teacher Education. Washington, D.C.: The Asso— ciation for Student Teaching, 47th Yearbook, 1969. Bischof, Ledford J. Interpreting Personality Theories. New York: Harper and Row, Publishers, 1964. Bock, R. Darrell, and Haggard, Ernest A. "The Use of Multivariate Analysis of Variance in Behavioral Research.” Handbook of Measurement and Assess— ment in Behavioral Sciences. Edited by Dean Whitla. Reading, Mass.: Addison—Wesley Pub. Co., 1968. Bowers, Norman D., and Soar, Robert S. Studies of Human Relations in the Teaching Learning Process: V: Final Report: Training for Classroom Teachers. Nashville, Tennessee: Vanderbilt University, 1961. Burdin, Joel L., and Lanzillotti, Kaliopee. A Reader‘s Guide to the Comprehensive Models for Preparing Elementary Teachers. Washington, D.C.: Eric Clearinghouse on Teacher Education, 1969. Cattell, Raymond B. "The Meaning and Strategic Use of Factor Analysis." Handbook of Multivariate Ex— perimental Psychology. Edited by Raymond B. Cattell. Chicago: Rand McNally & Co., 1966. 189 190 Cook, Walter W., Leeds, Carroll H., and Callis, Robert. "Predicting Teacher—Pupil Relations." The Eval— uation of Student Teaching. Twenty—eighth Year- book of the Association for Student Teaching. Washington, D.C.: The Association for Student Teaching, 1949. Corman, Bernard R., and Olmsted, Ann G. The Internship in the Preparation of Elementary School Teachers. East Lansing, Michigan: College of Education, Michigan State University, 1964. Cornell, F. G., Lindvall, C. M., and Saupe, J. L. Ap Exploratory Measurement of Individualities of Schools and Classrooms. Urbana: Bureau of Edu— cational Research, University of Illinois, 1952. Getzels, J. W., and Jackson, P. W. "The Teacher's Per— sonality and Characteristics." Handbook of Re— search on Teaching. Edited by N. L. Gage. Chicago: Rand McNally, 1963. Gordon, Ira J. Relationships Between Personality Vari- ables and Classroom Behavior of Teaching Interns. Gainsville, Florida: University of Florida, 1964. U.S. Department of Health, Education and Welfare, Project Number 1717. Guilford, J. P., and Lacy, J. I. Printed Classification Tests, Army Air Forces Aviation Psychological Program, Research Report 5. Washington, D.C.: Government Printing Office, 1947. Harman, Harry H. Modern Factor Analysis. Second edition, rev. Chicago: The University of Chicago Press, 1967. Hays, William L. Statistics for Psychologists. Chicago: Holt, Rinehart and Winston, 1963. Horst, Paul. "An Overview of the Essentials of Multi— variate Analysis Methods." Handbook of Multi— variate Experimental Psychology. Edited by ' Raymond B. Cattell. Chicago: Rand McNally & Co., 1966. 191 Katz, D., and Stotland, E. ”A Preliminary Statement to a Theory of Attitude Structure and Change." Vol. III of Psychology: A Study of a Science. Edited by S. Koch. Quoted in Insoko, Chester A. Theories of Attitude Change. New York: Meredith Publishing Co., 1967. Kerlinger, Fred N. Foundations of Behavioral Research. Chicago: Holt, Rinehart and Winston, Inc., 1964. Kirk, Roger E. Experimental Design: Procedures for the Behavioral Sciences. Belmont, California: Brooks/Cole Publishing Company, 1968. McNemar, Quinn. Psychological Statistics. Third edition. New York: John Wiley and Sons, Inc., 1962. Michigan State University, College of Education, EIP Elementary Intern Program, Another Way of Learn— ing to Teach. Final Report to the Ford Founda- tion, undated. Medley, Donald M., and Mitzel, Harold E. "Measuring Classroom Behavior by Systematic Observation.” Handbook of Research on Teaching. Edited by N. L. Gage. Chicago: Rand McNally, 1963. Murray, Henry A. Explorations in Personality. New York: Oxford Press, 1938. Pearson, Egon S., and Hartley, H. O. (eds.) Biometrika Tables for Statisticians, Vol. I. Cambridge, England: University Press, 1958. Reno, Raymond H. The Impact Teacher. St. Paul: 3M Education Press, 1967. Rokeach, Milton. The Open and Closed Mind. New York: Basic Books, Inc., 1960. Ryans, David G. Characteristics of Teachers. Washington, D.C.: American Council on Education, 1960. 192 Secord, Paul F., and Backman, Carl W. Social Psychology. New York: McGraw—Hill Book Company, 1964. Simon, Anita, and Boyer E. Gil (eds.) Mirrors for Be- havior. Philadelphia: Research for Better Schools, Inc., Classroom Interaction Newsletter, January, 1968. Snygg, Donald, and Combs, Arthur W. Individual Behavior. Rev. ed. New York: Harper & Row, Publishers, Inc., 1959. Verduin, John R., Jr. Conceptual Models in Teacher Edu— cation. Washington, D.C.: The American Associ— ation of Colleges for Teacher Education, 1967. Wallen, Norman E.; Travers, Robert M. W.; Reid, Ian E.; Iand Wodtke, Kenneth H. Measured Needs of Teachers and Their Behavior in the Classroom. Salt Lake City: University of Utah, 1961. Final Report to United States Department of Health, Education and Welfare, Office of Education Contract No. 444 (8029). Wilk, Roger E.; Edson, William H.; Davies, Don; and Chase, Naomi. A Study of the Relationship Between Ob- served Classroom Behaviors of Elementary Student Teachers, Predictors, of These Behaviorsz and Ratings by Supervisors. Minneapolis, Minnesota: University of Minnesota, 1962. Withall, John, and Lewis, W. W. "Social Interaction in the Classroom." Handbook of Research on Teaching. Edited by N. L. Gage. Chicago: Rand McNally & Co., 1963. Periodicals Anderson, Donald. ”Personality Attributes of Teachers in Organizational Climates." Jburnal of Educational Research, LXII (July, 1969), 441—443. 193 Barr, A. S. ”The Measurement and Prediction of Teaching Efficiency: A Summary of Investigations." Jour— nal of Experimental Education, XVI (June, 1948), 203—283. Bowers, Norman D., and Soar, Robert S. "The Influence of Teacher Personality on Classroom Interaction," Journal of Experimental Education, XXX (June, 1962), 309-311. Brown, Bob Burton, Mendenhall, William, and Beaver, Robert. "The Reliability of Observations of Teachers' Classroom Behavior." The Journal of Experimental Education, XXXVI (Spring, 1968), l—lO. Brumbaugh, R. B., Hoedt, Kenneth C., and Beisel, William H., Jr. "Teacher Dogmatism and Perceptual Ac— curacy." Journal of Teacher Education, XVII (Fall, 1966), 332—335. Callis, Robert. ”The Efficiency of the Minnesota Teacher Attitude Inventory for Predicting Interpersonal Relations in the Classroom." Journal of Applied Psychology, XXXVII (April, 1953), 82-85. Cappelluzzo, Emma M., and Brine, James. ”Dogmatism and Prospective Teachers.” Journal of Teacher Edu— cation, XX (Summer, 1969), 148—152. Cattell, Raymond B. ”The Three Basic Factor—Analytic Research Designs—~Their Interrelations and De— rivatives.” Psychological Bulletin, XLIX (Sep~ tember, 1952), 499—520. Cook, Desmond L., Linden, James D., and McKay, Harrison E. "A Factor Analysis of Teacher Trainee Responses to Selected Personality Inventories.” ‘and Psychological Measurement, XXI (Winter, 865-871. 1961), ”Attitude Changes of Beginning Teachers Journal of 326—328. Day, Harry P. . After Initial Teaching Experience.” Teacher Education, X (September, 1959), Educational 194 Domas, S. J., and Tiedman, D. V. ”Teacher Competence: An Annotated Bibliography.” Journal of EXperi- mental Education, XIX (December, 1950), 99—218. Eysenck, H. J. "The Logical Basis of Factor Analysis." American Psychologist, VIII (March, 1953), 105— 114. Frumkin, R. M. ”Dogmatism, Social Class, Values, and Academic Achievement in Sociology.” Journal of Educational Sociology, XXXIV (May, 1961), 398~403. Gordon, Ira J. "Assessment of Classroom Emotional Climate by Means of the Observation Schedule and Record.” Journal of Teacher Education, XVII (Summer, 1966), 224~232. Guilford, J. P. "When Not To Factor Analyze.” Psycho- logical Bulletin, XLIX (January, 1952), 26—37. Jackson, Philip W., and Guba, Egon G. ”The Need Structure of In—Service Teachers: An Occupational Analy— sis." School Review, LXV (June, 1957), 176—192. Jay, Rutledge L. "Q-Technique Factor Analysis of the Rokeach Dogmatism Scale." Educational and Psy— chological Measurement, XXIX (Summer, 1969), 453—459. Johnson, J. S. "Dogmatism: A Variable in the Prediction of Student Teaching Performances." Contemporary Education, XLI (October, 1969), 14—18. Kemp, C. Gratton. ”Improvement of Critical Thinking in Relation to Open—Closed Belief Systems." Journal of Experimental Education, XXXI (March, 1963), 321-323. Korn, Harold A., and Giddan, Norman S. ”Scoring Methods and Construct Validity of the Dogmatism Scale." Educational and Psychological Measurement, XXIV (Winter, 1964), 867—874. 195 Leeds, Carroll H. "A Second Validity Study of the Minne— sota Teacher Attitude Inventory." TElementary School Journal, LII (March, 1952), 396—405. Lehmann, I. J. ”Some Socio—Cultural Differences in Atti— tudes and Values." Journal of Educational So- ciology, XXXVI (September, 1962), 1—9. Lewis, W. W., and Newell, John M. "Analysis of Classroom Interaction Through Communication Behaviors." Journal of Experimental Education, XXX (June, 1962), 321—322. Linden, Kathryn W., and Linden, James D. "A Longitudinal Study of Teachers' Attitudes and Personality Characteristics." Journal of Teacher Education, XX (Fall, 1969), 351—360. Long, Barbara H., and Ziller, Robert C. ”Dogmatism and Predecisional Information Search." Journal of Applied Psychology, XLIX (October, 1965), 376—378. McGee, Henry M. "Measurement of Authoritarianism and its Relation to Teachers' Classroom Behavior." Gen— etic Psychology Monographs, LII (August, 1955), 89—146. MacLean, Malcolm 8., Jr. "Some Multivariate Designs for Communications Research.” Journalism Quarterly, XLII (Autumn, 1965), 614—622. Masling, Joseph, and Stern, George. "Effect of the Ob— server in the Classroom.” Journal of Educational Psychology, LX (October, 1969), 351-354. Medley, Donald M. "Teacher Personality and Teacher Pupil Rapport.” Journal of Teacher Education, XII (June, 1961), 152—156. Medley, Donald M., and Mitzel, Harold E. ”Some Behavioral Correlates of Teacher Effectiveness." Journal of Educational Psychology, L (December, 1959), 239— 246. 196 Medley, Donald M., and Mitzel, Harold E. "A Technique for Measuring Classroom Behavior." Journal of Educam tional Psychology, XLIX (April, 1958), 86-92. Medley, Donald M., and Mitzel, Harold E. ”A Tentative Framework for the Study of Effective Teacher Be— havior." Journal of Experimental Education, XXX (June, 1962), 317-320. Medley, Donald, Mitzel, Harold, and Rabinowitz, W. ”Long— itudinal Studies of a Group of Teacher Education Graduates.” Journal of Teacher Education, X (March, 1959), 117—119. Mitzel, Harold E., Rabinowitz, William, and Ostreicher, Leonard M. "The Effects of Response Sets on the Validity of the Minnesota Teacher Attitude Inven— tory." Educational and Psychological Measurement, XVI (Winter, 1956), 501-515. Mouw, John T. ”Effect of Dogmatism on Levels of Cognitive Processes.” Journal of Educational Psychology, LX (October, 1969), 365-369. Musella, D. ”Open—Closed~Mindedness as Related to the Ratings of Teachers by Elementary School Prin— cipals.” Journal of Expprimental Education, XXXV (Spring, 1967), 75—79. Ohnmacht, F. W. ”Teacher Characteristics and Their Rela— tionship to Some Cognitive Styles." Journal of Educational Research, LX (January, 1967), 201—204. Poe, Charles A. "Convergent and Discriminant Validation of Measures of Personal Needs." Journal of Edu— cational Measurement, LXI (Summer, 1969), 103—107. Rabkin, Leslie Y. "The Dogmatism of Teachers?“ Journal of Teacher Education, XVII (Spring, 1966), 47—49. Rokeach, Milton. ”The Nature and Meaning of Dogmatism.“ Psychological Review, LXI (May, 1954), 194-204. 197 Rokeach, Milton. “Political and Religious Dogmatism: An Alternative to the Authoritarian Personality." Psychological Monographs: General and Applied, LXX (Number 18, Whole Number 425, 1956), 1—43. Scandrette, Onas. "Differential Need Patterns of Women Elementary and Secondary Level Student Teachers." Journal of Educational Research, LV (May, 1962), 376—379. Seibel, D. W. ”Predicting the Classroom Behavior of Teachers." Journal of Experimental Education, XXXVI (Fall, 1967), 26-32. Sheldon, M. S., Coale, J. M., and Copple, R. "Current Validity of the “Warm Teacher Scales.'" Journal of Educational Psychology, L (February, 1959), 37—40. Silberman, Melvin L. ”Behavioral Expression of Teachers' Attitudes Toward Elementary School Students." Journal of Educational Psychology, LX (October, 1969), 402-407. Smith, B. Othanel. "Recent Research on Teaching: An Interpretation.” The High School Journal, LI (November, 1967), 63h74. Soderbergh, P. A. ”Dogmatism and the Public School Teacher." Journal of Teacher Education, XV (September, 1964), 245—251. Stephenson, William. ”The Inverted Factor Technique.” British Journal of Psychology, XXVI (April, 1936), 344—361. Stephenson, William. "Some Observations on Q Technique." Psychological Bulletin, XLIX (September, 1952), 483—498. HP* 198 Teigland, John J. "Relationship Between Measured Teacher Attitude Change and Certain Personality Charac— teristics." Journal of Educational Research, LX (October, 1966), 84—85. Troldahl, Verling C., and Powell, Fredric A. "A Short— Form Dogmatism Scale for Use in Field Studies." Social Forces, XLIX (December, 1965), 211—214. Wallen, Norman E.; Travers, Robert M. W.; Reid, Ian E.; and Wodtke, Kenneth H. ”Relationships Between Teacher Needs and Teacher Behavior in the Class— room.” Journal of Educational Psychology, LIV (February, 1963), 23-32. Withall, John. "Development of a Technique for the Mea— surement of Socio—Emotional Climate in Class— rooms.” Journal of Experimental Education, XVII (March, 1949), 347—361. Tests and Manuals Cook, Walter W., Leeds, Carroll H., and Callis, Robert. Minnesota Teacher Attitude Inventory, Form A. New York: The Psychological Corporation, 1951. Cook, Walter W., Leeds, Carroll H., and Callis, Robert. Minnesota Teacher Attitude Inventory Manual. Rev. ed. New York: The Psychological Corpor— ation, 1959. Earlier edition, 1951. Edwards, Allen L. Edwards Personal Preference Schedule. New York: The Psychological Corporation, 1953. Edwards, A. L. Edwards Personal Preference Schedule: Manual. Rev. ed. New York: The Psychological Corporation, 1959. Rokeach, Milton. The Dogmatism Scale. Short form-— mimeographed. 199 Unpublished Materials Bridgman, John Northan. ”Selected Teacher Characteristics and Their Relationships with Certain Behavior Pat— terns and Teaching Effectiveness.” Unpublished Ed.D. dissertation, The University of North Caro— lina at Chapel Hill, 1967. Dissertation Abstracts, XXVIII (March, 1968), 3524 A. Clary, Eldon Gandy. "Predicting Student Teaching Behavior From Needs Profiles by Comparison with Sociomet— rically Defined Groups." Unpublished Ed.D. dis— sertation, North Texas State University, 1968. Dissertation Abstracts, XXIX (April, 1969), 3488— 3489 A. Ferguson, John L., Jr. ”A Factorial Study of the Minne— sota Teacher Attitude Inventory.“ Unpublished Ed.D. dissertation, University of Missouri, 1958. Dissertation Abstracts, XIII (No. 6, 1953), 1087. Friedman, L. R. ”An Investigation of Certain Teaching Behaviors of and Influences Upon New Elementary Teachers.” Unpublished Ed.D. dissertation, The University of Florida, 1967. Dissertation Ab- stracts, XXIX (July, 1968), 165 A. Furr, Oneta Roberts. "Effects of Observational Feedback on Verbal and Nonverbal Classroom Behavior of Student Teachers.” Unpublished Ed.D. disserta— tion, North Texas State University, 1968. Dis— sertation Abstracts, XXIX (April, 1969), 3495 A. Johnson, James. "The Relationship of Open— and Closed— Mindedness to Success in Student Teaching." Un— published Ed.D. dissertation, George Pebody Col- lege for Teachers, 1966. Jones, Lyle. "Some Illustrations of Psychological EXperi_ ments Designed for Multivariate Statistical Anal— ysis." Paper presented at the conference on Multivariate Experimentation, Allerton Park, Uni— versity of Illinois, November 15, 1960. .- 200 Leeds, Carroll H. "The Construction and Differential Value of a Scale for Determining TeacherFPupil Attitudes." Unpublished doctoral dissertation, University of Minnesota, 1946. American Psy— chologist, III (July, 1948), 296-297. Lewis, Franklin Garner. ”The Relationship of Authori- tarianism as Revealed by the Rokeach Dogmatism Scale and Perceived Effectiveness of Teaching as Indicated by Teachersu Self—Ratings, PrincipalsI Ratings and Supervisorsu Ratings." Unpublished Ed.D. dissertation, North Texas State University, 1968. Dissertation Abstracts (December, 1968), 1682 A. Oana, Robert G. "An Analysis of the Use of the Minnesota Teacher Attitude Inventory in a Pre—service Pro— gram in Childhood Education.” Unpublished Ed.D. dissertation, Columbia University, 1965. Piele, Philip Kern. ”The Relationship of Teacher Open and Closed Mindedness to Classroom Verbal Be- havior.” Unpublished Ph.D. dissertation, Univer— sity of Oregon, 1968. Dissertation Abstracts, XXIX (May, 1969), 3879—3880 A. Ragsdale, Elva Mae. "Attitude Changes of Elementary Stu— dent Teachers and the Changes in Their Classroom Behavior During Student Teaching." Unpublished Ed.D. dissertation, Ball State University, 1967. Dissertation Abstracts, XXVIII (August, 1967), 521—522 A. Rippy, Mark Leo, Jr. "Certain Relationships Between Classroom Behavior and Personality Characteristics of Selected Elementary Teachers." Unpublished Ed.D. dissertation, George Peabody College for Teachers, 1960. Samph, Thomas. ”Observer Effects on Teacher Behavior." Unpublished Ph.D. dissertation, University of Michigan, 1968. Dissertation Abstracts, XXIX (February, 1969), 2573 A. Southworth, Horton C. "A Study of Certain Personality and Value Differences in Teacher Education Majors Pre- ferring Early and Later Elementary Teaching ' Levels." Unpublished Ed.D. dissertation, Michigan State University, 1962. APPENDICES APPENDIX A THE MANIFEST NEEDS ASSOCIATED WITH EACH OF THE FIFTEEN EDWARDS PERSONAL PREFERENCE SCHEDULE VARIABLES APPENDIX A THE MANIFEST NEEDS ASSOCIATED WITH EACH OF THE FIFTEEN EDWARDS PERSONAL PREFERENCE SCHEDULE VARIABLES* l. Achievement — To do one's best, to be success— ful, to accomplish tasks requiring skill and effort, to be a recognized authority, to accomplish something of great significance, to do a difficult job well, to solve diffi— cult problems and puzzles, to be able to do things better than others, to write a great novel or play. 2. Deference — To get suggestions from others, to find out what others think, to follow instructions and do what is expected, to praise others, to tell others that they have done a good job, to accept the leadership of others, to read about great men, to conform to custom and avoid the unconventional, to let others make decisions. 3. Order — To have written work neat and organ- ized, to make plans before starting on a difficult task, to have things organized, to keep things neat and orderly, to make advance plans when taking a trip, to organize de— tails of work, to keep letters and files according to some system, to have things arranged so that they run smoothly without change, to have meals organized and a definite time for eating. 4. Exhibition — To say witty and clever things, to tell amusing jokes and stories, to talk about personal '*A.”L. Edwards, Edwards Personal Preference Schedule: ‘Manual (New York: ,The=Psychological Corporation, 1953), 11. 204 adventures and experiences, to have others notice and comment upon one's appearance, to say things just to see what effect it will have on others, to talk about per— sonal achievements, to be the center of attention, to use words that others do not know the meaning of, to ask questions others cannot answer. 5. Autonomy — To be able to come and go as de— sired,to say what one thinks about things, to be indepen— dent of others in making decisions, to feel free to do what one wants, to do things that are unconventional, to avoid situations where one is expected to conform, to do things without regard to what others may think, to crit— icize those in positions of authority, to avoid respon— sibilities and obligations. 6. Affiliation — To be loyal to friends, to par— ticipate in friendly groups, to do things for friends, i to form new friendships, to make as many friends as pos— sible, to share things with friends, to do things with friends rather than alone, to form strong attachments, I to write letters to friends. 7. Intraception — To analyze one's motives and feelings, to observe others, to understand how others feel about problems, to put one's self in another's place, to judge people by why they do things rather than by what they do, to analyze the motives of others, to predict how others will act. 8. Succorance — To have others provide help when in trouble, to seek encouragement from others, to have others be kindly, to have others be sympathetic and under— standing about personal problems, to receiveaa great deal of affection from others, to have others do favors cheer— fully, to be helped by others when depressed, to have others feel sorry when one is sick, to have a fuss made over one when hurt. 9. Dominance — To argue for one's point of view, to be a leader in groups to which one belongs, to be re— garded by others as a leader, to be elected or appointed chairman of committees, to make group decisions, to settle arguments and disputes between others, to persuade and 206 influence others to do what one wants, to supervise and direct the actions of others, to tell others how to do their jobs. 10. Abasement — To feel guilty when one does something wrong, to accept blame when things do not go right, to feel that personal pain and misery suffered does more good than harm, to feel the need for punishment for wrong doing, to feel better when giving in and avoid— ing a fight than when having one's own way, to feel the need for confession of errors, to feel depressed by in- ability to handle situations, to feel timid in the pres— ence of superiors, to feel inferior to others in most respects. ll. Nurturance ~ To help friends when they are in trouble, to assist others less fortunate, to treat others with kindness and sympathy, to forgive others, to do small favors for others, to be generous with others, to sympathize with others who are hurt or sick, to show a great deal of affection toward others, to have others confide in one about personal problems. 12. Change — To do new and different things, to travel, to meet new people, to experience novelty and change in daily routine, to experiment and try new things, to eat in new and different places, to try new and dif— ferent jobs, to move about the country and live in dif— erent places, to participate in new fads and fashions. l3. Endurance — To keep at a job until it is finished, to complete any job undertaken, to work hard at a task, to keep at a puzzle or problem until it is solved, to work at a single job before taking on others, to stay up late working in order to get a job done, to put in long hours of work without distraction, to stick at a problem even though it may seemzas if no progress is being made, to avoid being interrupted while at work. 14. Heterosexuality — To go out with members of the opposite sex, to engage in social activities with the opposite sex, to be in love with someone of the oppo— site sex, to kiss those of the opposite sex, to be re— garded as physically attractive by those of the opposite 207 sex, to participate in discussions about sex, to read books and plays involving sex, to listen to or to tell jokes involving sex, to become sexually excited. 15. Aggression — To attack contrary points of view, to tell others what one thinks about them, to crit— icize others, to make fun of others, to tell others off when disagreeing with them, to get revenge for insults, to become angry, to blame others when things go wrong, to read newspaper accounts of violence. APPENDIX B ROKEACH DOGMATISM SCALE (SHORT FORM) DIRECTIONS: APPENDIX B ROKEACH DQfiMATISM SCALE (SHORT FORM) You will read below some statements people have made as their opinion on several topics. You may find yourself agreeing strongly with some of the statements, disagreeing just as strongly with others, and perhaps uncertain about others. Whether you agree or disagree with any statement, you can be sure that many other people feel the same as you do. We want your personal Opinion on each state— ment. When you read each one, first indicate whether, in general, you agree or disagree with it: In column I mark (+) or (e): + = agree — = disagree Then indicate how strongly you agree or disagree: 1p column I; mark 1, 2, or 3: Agree Disagree 1. Agree a little 1. Disagree a 2. Agree on the whole little 3. Agree very much 2. Disagree on the whole 3. Disagree very much In this complicated world of ours the only way we can know what's going on is to rely on leaders or experts who can be trusted. My blood boils whenever a person stubbornly refuses to admit he's wrong. 209 10. 11. 12. 13. 14. 15. 16. 17. There are two kinds of people in this world: those who are for the truth and those who are against the truth. Most people just don't know what's good for them. Of all the different philosophies which exist in this world there is probably only one which/is correct. The highest form of government is a democracy and the highest form of democracy is a gov— ernment run by those who are most intelli— gent. The main thing in life is for a person to want to do something important. I'd like it if I could fine someone who would tell me how to solve my personal problems. Most of the ideas which get printed nowadays aren't worth the paper they are printed on. Man on his own is a helpless and miserable creature. It is only when a person devotes himself to an ideal or cause that life becomes meaning— ful. Most people just don't give a "damn" for others. To compromise with our political Opponents is dangerous because it usually leads to the betrayal of our own side. It is often desirable to reserve judgment about what's going on until one has had a chance to hear the opinions of those one re- spects. The PRESENT is all too often full of unhap— piness. It is only the FUTURE that counts. The United States and Russia have just about nothing in common. In a discussion I often find it necessary to repeat myself several times to make sure I am being understood. 18. 19. 20. While I don't like to admit this even to myself, my secret ambition is to become a great man, like Einstein, or Beethoven, or Shakespeare. Even though freedom of speech for all groups is a worthwhile goal, it is unfortunately necessary to restrict the freedom of certain political groups. It is better to be a dead hero than to be a live coward. APPENDIX C OBSERVATION SCHEDULE AND RECORD (OSCAR III) MANUAL AND RECORDING FORM APPENDIX C OBSERVATION SCHEDULE AND RECORD (OSCAR III) MANUAL AND RECORDING FORM Historical Development Originally, the Observation Schedule and Record (OScAR) was designed by Medley and Mitzel to provide quantitative data regarding classroom behaviors of begin— ning teachers. Items constructed by Cornell, Lindvall, and Saupe (1952) and Withall (1949) were modified and combined to form OScAR 2a. This classroom observation instrument differs from its predecessors in three ways: 1. Observer accuracy is increased. Specific behav- ioral cues requiring little or no judgment on the part of the observer are employed. Hence rela— tively untrained observers can use the OSCAR easily. Observations are made by a single observer visit— ing a classroom by himself. A score based on simultaneous observations by two observers of a teacher contains less information than an average score based on observations mafie by two observers who see a teacher at different times. In the latter case, the behavior sample is doubled. The process of scoring is separated from the pro— cess of observing. The observer is asked to re— cord any possibly significant behaviors that occur; he does not judge their relationship to any dimension or scale. 213 214 OScAR III includes items concerning teacher and pupil activities and relationships, grouping, materials used, and subject content discussed during three separate five— minute intervals. Fifteen groups (lettered A—O) of cate— gories appear on the recording sheet. Behaviors are recorded in columns I, III, and V during the corresponding first, third, and fifth five- minute periods in a twenty—five minute observation. Immediately after recording the necessary identification data, the observer starts his stOpwatch and checks in the appropriate row and column each different activity or behavior which occurs during the particular observation interval. Note that any category on this observation sheet is checked only once for each five—minute period regardless of the number of times the activity or behavior occurs during the observation period. Definition and Explanation of Categories E Group A: Teacher-Pupil 1. Teacher questions, pupil answers. The teacher's question may be either direct or implied. However, it must be a question for which an answer is expected, that is, the teacher must pause long enough for an apprOpriate answer to be given. The question cannot be rhetorical. A teacher calls on a pupil, points to a pupil, or in another way indicates that a pupil is to respond. A statement such as, ”I don't know," or an incorrect answer, in this case, constitutes an answer. If the pupil ignores the question, category 4 of group D should be checked. This category differs from cate- gories l, 2, and 3 of group E in that, in this category, the questions require only brief answers or comments (thus different from El and E3) which are pp; dependent upon memorization by rote (thus different from E2). Some examples of questions in this category are, "Billy, did you pass your paper to the front of the room," "Jane, which spelling word did you have the most 215 difficulty learning," or, "Bob, do you understand how Jimmy worked this problem." 2. Teacher answers pupil's question. As in Al, the question may be either direct or implied, however, it must be a question for which an answer is expected (see Al). If the teacher ignores the question, category 4 of group A should beichecked. 3. Teacher interrupts pupil. This category is checked when the teacher speaks while a pupil is speaking regardless of the reason for the interruption, e.g., for clarification, additional information, change of tOpic, or correction. 4. Teacher igpores pupil's question. This cate— gory is checked whenever the teacher, either deliberately or inadvertently, ignores, or does not answer or comment on a question by a pupil, provided that the pupil expects an answer to the question (see Al) and that the question is asked in such a way that it is, in all probability, L audible to the teacher. This category is checked regard— ‘ less of the occasion during which the question is asked, for example, during a test. 5. Teacher warns or threatens pupil or pupils. This must be a verbal threat or warning and the threat or warning must be explicit in the statement of the teacher. A threat or warning would ordinarily take the form, "If A does (doesn't) do X then Y (will happen)." More specifically, ”If you don‘t get the problems fin— ished before class is dismissed you will stay after school." 6. Teacher uses criticigm and/or sarcasm. This category includes scorn, ridicule, or (contemptuous) laughter, but the statement or the laughter must be such that the content, the context, or the manner of expres— sion explicitly indicates or portrays criticism, sarcasm, scorn, or ridicule. Academic correction, e.g., three times four is twelve not nine are ppp_included in this category unless criticism for not knowing the correct answer, or sarcasm, accompanies the correction. 216 7. Teacher praises and/Cr encourages pupil or pupils forpgood work. The praise or the encouragement may take the form of a comment or a remark to the indi— vidual concerned, e.g., "This is the best arithmetic paper you have submitted this week"; to the class about the individual, e.g., "If the rest of the class did their problems as well as X. . ."; to a specified part of the . class, e.g., "Those of you who solved problem five cor— rectly have learned the multiplication tables well”; or to the entire class, e.g., ”All of you must have studied very hard to have learned the multiplication tables so well in such a short time.” This category is checked when the praise or the encouragement is for a task or an activity well done. 8. Teacher praises andgor encourages pupil or pupils for good behavior. This category is the same as A7 except that the praise or encouragement is for good behavior rather than for good work. Thus, the praise or encouragement may take the form of a comment or a remark to the individual concerned, to the class about the in— ‘ dividual, to a specified part of the class, or, to the entire class. 9. Teacher works with individual pupil. This category is checked when the teacher aids an individual pupil regardless of whether the aid is asked for by the pupil and regardless of whether the aid is of an "aca— demic” nature or not (it might, for example, be helping a pupil put on boots or fix a toy). Group B: Teacher—Class 1. Teacher lectures. This category is checked during the time when the class is conducted in a manner which makes no provision for active pupil participation in the lesson being presented except through some form of intrusion or interruption in the teacher's presenta— tion. If an interruption occurs or should active par— ticipation on the part of the pupils be initiated then another category is to be checked (see A2, A4, or B3). 3. 217 2. Teacher reads, tells story. This category is checked whether the story is told or the passage read by the teacher to clarify or to further elaborate a partic— ular point in a lecture or whether the story or the read— ing itself is the primary concern of the teacher and the class at the particular time. 3. Teacher talks to class. This category is checked when the teacher is discussing topics of a non— academic nature, for example, running in the hall, good sportsmanship, obeying the school patrol; if the teacher is giving directions to the class, e.g., "Turn to page 41 of your arithmetic books," or when the teacher is answering or asking questions so:that there is intermit— tent pupil participation during the imparting of infor— mation. 4. Teacher illustrates at board. This category is checked when the teacher uses the blackboard as a means of facilitating explanation, elaboration, clarification, justification, or the imparting of information. 5. Teacher illustrates with map or chart. This category includes the use by the teacher of graphs, pic— tures, posters, or other pictorial material and is checked when one of these visual aids is used by the teacher as a means of facilitating explanation, elaboration, Clarifica— tion, justification, or the imparting of information. 6. Teacher demonstrates. This category is checked when the teacher uses a three—dimensional model, object, or specimen as a means of facilitating explana— tion, elaboration, clarification, justification, or the imparting of information. 7. Teacher shows film, slides, plays records. This category is checked when the teacher for any reason whatsoever, utilizes these particular audio, visual, or audiovisual materials. 8. Teacher passes books, paper, milk. This cate- gory.is checked whenever the teacher is involved in the distribution or the passing of a book or books, paper, or milk. I i .- 218 9. Teacher leads singing, exercises, or games. This category is checked when the teacher is actively engaged as the director of singing, drill exercise, or games. Group C: Teacher Activities 1. Teacher works at desk (or at a temporary lo— cation serving as a desk). This category is checked at those times when the teacher is engaged in making out reports, grading papers, composing lesson plans, and similar tasks. Should the teacher temporarily stop this activity in order to help a pupil or pupils, a check should be made in A9 or in some category of Group I or J. 2. Teacher cleans, decorates room. This cate- gory is checked when the teacher is actively engaged in cleaning (picking up paper, dusting) or decorating the room. 3. Teacher writes on, decorates board. This category is checked whenever the teacher, while the class is busy at some other task, decorates or writes on or posts information on a bulletin board or blackboard. This category is distinct from, and is not to be confused with category 4 of Group B. 4. Teacher leaves, enters room. This category is checked when the teacher leaves or enters the room except for the time when the teacher stands in the door— way or in the hall within sight and hearing of the class. 5. Teacher talks with visitor. This category is checked when the teacher talks to a visitor in the classroom, in the doorway, or in the hall adjacent to the classroom. Group D: Pupil—Teacher l. Pupil whispers. This category is checked only when the whiSpering on the part of the pupil indicates or reflects, by the time and the place of its occurrence, .1._._._‘.. 19+. 219 inattention on the part of a pupil or pupils. Thus, this category is checked only when the whispering occurs while the teacher is eXpecting attention. 2. Pupil laughs. This category is checked only when the laughter is audible and when it indicates or reflects, by the time and the place of its occurrence, inattention on the part of a pupil or pupils. Laughter with the class at an amusing story would, for example, not be included in this category while laughter resulting from the telling of a "private joke" would be included in this category. 3. Pupil interrupts teacher. This category is checked when the pupil speaks while the teacher is speak— ing regardless of the reason for the interruption, e.g., for clarification, additional information, change of tOpic, or correction. 4. Pupil ignores guestion of teacher. This cate— gory is checked whenever the pupil, either deliberately or inadvertently, ignores, or does not answer or comment on, a question of the teacher provided that the teacher expects an answer to the question (see Al) and that the question is asked in such a way that it is, in all prob— ability, audible to the pupil. This category is checked regardless of the occasion during which the question is asked, for example, during a test or a play period. 5. Pupil shows hostility toward teacher. This behavior may be either verbal or non—verbal, but it must be such as to clearly indicate, not merely imply, hos— tility. The hostility may be in the form of a direct re— fusal, a definite resistance, or a distinct reluctance on the part of the pupil to perform some taSk or to take some particular course of action; it may be a specific comment of a derogatory or derisive nature made to the teacher or to another pupil about the teacher, or it may take the form of an expression or gesture made by the pupil to the teacher or behind the teacher's back, e.g., to stick out his tongue or to ”make a face“ at the teacher. This category would also be checked should the pupil hit, or strike at, or throw some object at the teacher. L 220 6. Pupil shows affection for teacher. This bee havior may be either verbal or non—verbal, but it must be such as to clearly indicate, not merely imply, affec— tion for the teacher. The affection may be expressed in the form of a comment or remark to the teacher or to another pupil about the teacher, e.g., ”You are my fav— orite teacher,” ”You are nice,“ ”I like Miss X," or it may be a direct expression of affection, i.e., putting his arm around the teacher or taking the teacher's hand. Group E: Pupil-Class l. Pupil talks to class or group. This category is checked when a pupil talks extemporaneously, except for those times when the pupil simply gives brief answers to a question (see Al). This category is checked when, for example, a pupil tells what he saw on the way to school, what he liked best at the fair, or what he thought of a story. 2. Pupil recites. This category is checked when the pupil presents material learned by rote, usually in the form of a memorized poem or story or a particular fact or set of facts such as, for example, "What states border Michigan," ”Name the thirteen colonies,” ”Who in— vented the cotton gin.“ 3. Pupil reports, gives prepared talk. This category is checked when the pupil gives an organized prepared report or talk whether from memory (not to be confused with E1), read from a paper, or given with the aid of note or reference cards. 4. Pupil reads aloud. This category is checked when the pupil, usually at the request of the teacher, reads aloud from the text, a workbook, or similar mate- rial. 5. Pupil demonstrates, illustrates. This cate— gory is checked whenever the pupil uses the blackboard, a three-dimensional model, a map, chart, graph, picture, poster, or other pictorial material as a means of facil- itating explanation, elaboration, clarification, or the Note that this category is Pupil works at board. imparting of information. different and distinct from F4: This category is 6. Pupil gives skit or play. checked when the pupil takes an active part in the pre- sentation of a skit or a play. 7. Pupil sings, plays instrument. This category is checked whenever a pupil sings or plays an instrument whether this is done alone or in a group. 8. Pupil plays game. This category is checked whenever two or more pupils play a "recreational” or an ”educational” game. If only one person is involved, such as working a puzzle, F2 is checked. 9. Pupil leads class. This category is checked whenever a pupil assumes the reSponsibility for a class presentation or a class function of an academic nature, i.e., excluding singing, exercise, games (see E10). In this category, the pupil ordinarily operates as a class chairman, class president, panel moderator or member, or in a similar capacity. This category is different and distinct from El, E2, E3, and E4 for in these categories the teacher remains the active class leader despite the pupil contribution. 10. Pupil leads singing, exprcise, games. This category is checked whenever a pupil leads the class in singing, exercises, or games. This cate— 11. Pupil passes book(s), paper, milk. gory is checked whenever a pupil, or pupils, with or withe out the consent or the aid of the teacher, performs the specified or similar activities. Group F: Pupil Activities Pupil reads, studies at seat. This category 1. is checked whenever the pupil reads or studies at his or at a location which is being temporarily used F2 desk, as a desk. If the pupil is both reading and writing, is checked. 222 2. Pupil writes, manipulates at seat. This cate- gory is checked whenever a pupil does any writing or cal— culating or manipulating of some object, for example, an abacus, clock model, or flash card at his desk, or at a location which is being temporarily used as a desk. is, however, not checked when the pupil is doing art or This (see F3). 3. Pupil paints, draws, cuts, pastes at seat. This category is checked whenever a pupil is engaged in some type or kind of art or craft work, either at his desk or at some location used by the pupil as a temporary _craft work desk. This category is 4. Pupil works at board. checked whenever a pupil uses the blackboard, either at his own discretion or at the direction of the teacher, except when the pupil uses the blackboard as a means of facilitating explanation (see E5) or when the pupil dec— orates the board (see F5). 5. Pupil decorates room, board. This category is checked whenever a pupil, with or without the direction or permission of the teacher, decorates a bulletin board, or wall. blackboard, window, This category is 6. Pupil cleans room, board. checked Whenever a pupil, with or without the direction erases or washes the black— or request of the teacher, board, picks up paper from the floor, or performs similar tasks. 7. Pupil rests, has snack. This category is checked whenever a pupil, with or without permission of the teacher, rests on a mat, or at his seat, or has a snack. 8. Pupil leaves room. This category is checked whenever a pupil, with or without permission of the teacher, leaves the classroom. 223 Group G: Pupil-Pupil l. Pupil scuffles, fights with pupil. This cate— gory is checked whenever two or more pupils engage in hitting, slapping, wrestling, or some other form of scuffling or fighting. This category is thus checked only when force or contact is present. 2. Pupil shows hostility toward pupil. This category is checked whenever a pupil throws some object, for example, paper, eraser, pencil, or paper clip at another pupil; when a pupil directly threatens or warns another pupil, i.e., "If X (behavior) continues or occurs theanwfill happen (as a result of his action)," or when the threat or warning is explicit in the action of the pupil, e.g., shaking a fist at a pupil; or whenever a pupil uses words which, in their context, indicate anger, bitterness, scorn, ridicule, sarcasm, rebuff, or deri— sion of, or for, toward, or about, another pupil. Thus, the hostility may be either verbal, or nonverbal, or both. It may be direct or indirect, i.e., it may be to or about | the pupil concerned. However, it must be clearly indi— cated hostility by the action, the expression, the context, or by all three. .___4;v 3. Pupil shows affection toward pupil. This category is checked when a pupil talks or laughs with another pupil, when he takes another pupil's hand, puts his arm around another pupil, smiles at another pupil, or makes some remark such as, "I like you,” or llI like X" to or about another pupil. Thus, the expression of affection may be either verbal or nonverbal or both. However, it must be a clear expression of affection by the action, the expression, the context, or by all three. Group H: Grouping 1. Teacher choice of volunteers. This category is checked when the teacher makes the choice of persons to be part of a particular group, or to perform some activity or task entirely from among those pupils who volunteered (by raising their hand or some other method) for the task, or activity, or group. Note: If the selection is based in part on volunteers and in part on some other basis, H2 should be checked. 2. Teacher chéice\on some arbitrary"baSis. -This category is checked when the teacher makes the choice of persons to be part of a particular group, or to perform some activity or task, partially or entirely on some basis other than pupils who volunteered, e.g., "All those that missed problem three come to the front of the room." If it is not certain whether the pupils volunteered for the particular group, or activity, or task, for example, "Now that we have finished the spelling lesson you may re—form your social studies committees and continue to work on your special group problems," this category is to be checked. 3. Pupil choice of volunteers. This category is checked when a pupil, as a class or group leader, is re— sponsible for and makes the selection of other pupils for a particular group, or activity, or task, entirely from among those pupils who volunteered (by raising their hands or by some other method) for the group, or activity, or task. 4. Pupil choice on some arbitrary basis. This category is checked when a pupil, as a class or group leader, is responsible for and makes the selection of other pupils for a particular group, or activity, or task, in whole, or in part, on some basis other than volunteers. 5. Pppil goes to the group of his choice. This category is checked whenever the pupils are free to go to any group of their own choosing, for example, "All of you who wish to help with a report on George Washington go to the table in the back of the room, those of you>who wish to help with a report on Abraham Lincoln can use my desk, and those of you who wish to help make a report on Woodrow Wilson can use the table at the side of the room." 225 Group I: Individual task with teacher. A task is an indi— vidual task when a pupil is engaged in some activity or project in which the pupil alone is responsible for de- termining and/br accomplishing a goal and/or the means for attaining that goal. "With teacher" means that the teacher is not merely present in the room but that he is engaged in aiding an individual pupil or pupils. Il, 12, or 13 is checked depending upon the size of the group, i.e., over one half the class,four pupils to half the class, or two or three pupils. Group J: Group task with teacher. A task is a group task when two or more pupils are engaged in some activity or project in which the pupils together--through discussion and planning~—are responsible for determining and/or ac- complishing the goal and/or the means for attaining that goal. It should be noted that a category in this group is pp; checked when two or more pupils are performing the same task individually. The remainder of the distinctions made in this group are the same as in Group I. Groups K and L: Individual task without teacher. "Without teacher" means that the teacher is not engaged in aiding an indi— vidual pupil or pupils. It should be noted that some category in Groups K or L (Groups "without teacher") would be checked when, for example, the teacher is in the c1ass~ room but is grading papers or doing some other ”desk" work. The remainder of the distinctions which are neces— sary to make regarding the categories of Groups K and L have been discussed in connection with Groups I and J. 226 Groups M and N: These groups pertain to the materials used by the teacher and the pupils respectively in carrying out their activities. The materials included are: 1. blackboard 2. map, chart, or picture 3. slide, film strip, or film 4. audio aid (such as tape recorder, microphone, or record) 5. three dimensional object (such as model) 6. textbook or workbook 7. supplementary reading materials 8. writing materials 9. handicraft or art materials 0. special teaching or learning materials not itemized 11. no material used If special teaching or learning materials which are not itemized are used by the teacher and/or the pupils, the observer should specify at the bottom of the appropriate group the\particular material or materials being used. Group 0: This group pertains to the content area or areas which are discussed by the teacher and the pupils during the observation period. The content area on which the teacher spends the majority of his time during the ob- servation period should be checked twice. If ”test” is checked, some other category or categories pertaining to the subject area or areas the pupils are being tested on should also be checked. NOTE: The definitions of categories for OScAR III were taken from the USOE research project conducted by Roger E. Wilk, William H. Edson, Don Davies, and Naomi C. Chase, A Study of the Relationship Be— tween Observed Classroom Behaviors of Elementary 227 Student Teachers, Predictors of Those Behaviors, and Ratings by Supervisors (University of Minne— sota, 1962), Appendix A-3. Minor revisions in the text were made. 228 OZmGu. ~Z~u>1 Umo>0 _o ”nun” HHHHH HHHHH HHHHH HHHHH HHHHH HHHHH HHHHH HHHHH HHHHM HHHHH HHHHH unnnn MMHHH HHHHH HHHHH HHHHH I'll II I |I|.| wflo P N > H H +Crr llll when, at p ta 9.; m m wcfldsono .9 p Q #09 > HHH H HHH mnomno horomoe APPENDIX D MEANS AND STANDARD DEIVATIONS FOR VARIABLES ON THE OBSERVATION SCHEDULE AND RECORD AND PERSONALITY VARIABLES 230 Table A.l.——Means and Standard Deviations of OScAR Cate— gories: Pupil Activitiesa category Mean S.D. Pupil—Teacher Pupil whispers 5.43 4.66 Pupil laughs 1.64 2.35 Pupil interrupts teacher 4.66 3.45 Pupil ignores question of teacher 0.72 1.27 Pupil shOWS hostility to teacher 0.99 1.70 Pupil shows affection to teacher 0.84 2.00 Pupil-Class Pupil talks to group 3.31 3.13 Pupil recites 3.97 3.64 Pupil reports, gives prepared talk 0.37 0.87 Pupil reads aloud 3.66 2.78 Pupil demonstrates, illustrates 1.36 1.72 Pupil gives skit, play 0.23 0.68 Pupil sings, plays instrument 0.32 0.88 . Pupil plays game 0.81 1.44 ‘ Pupil leads class 0.73 1.27 Pupil leads singing, exercises, game 0.25 0.68 Pupil passes books, paper, milk 1.41 1.46 Individual Pupil Pupil reads, studies at seat 5.66 4.03 Pupil writes, manipulates at seat 8.56 4.35 Pupil paints, cuts, draws 2.38 2.45 Pupil works at board 1.00 1.56 Pupil decorates room, board 0.27 0.71 Pupil cleans room, board 0.66 1.24 Pupil rests, has snack 0.58 1.12 3.77 2.81 Pupil leaves, enters room Pupilepupil Pupil scuffles, fights with pupil 0.33 0.78 Pupil shows hostility to pupil 1.59 1.88 1.16 2.02 Pupil shows affection to pupil aBased on eighteen observations of 180 teachers. 231 Table A.2.——Means and Standard Deviations of OScAR Cate- gories: Teacher Activitiesa Category Mean S.D. Teacher—Pupil Teacher questions, pupil answers 11.34 3.51 Teacher answers pupil question 8.37 4.64 Teacher interrupts pupil 1.41 1.79 Teacher ignores pupil question 1.14 1.95 Teacher warns, threatens pupil 3111 3.07 Teacher criticizes, uses sarcasm 3.31 3.44 Teacher praises good work 4127 3.90 Teacher praises good behavior 1.44 2.04 Teacher works with individual pupil 6.27 3.72 Teacher—Class Teacher lectures 1.58 2.37 Teacher reads, tells story 1.16 1.41 Teacher talks to class 11.27 3.42 Teacher illustrates at board 3-25 2.55 Teacher illustrates with map, chart 0.61 0.95 Teacher demonstrates 1.60 1.84 Teacher shows film, slide 0.61 1.25 Teacher passes books, paper, milk 1.22 1.44 Teacher leads singing, exercises, game 0.68 1.34 Solitary Teacher works at desk 1.48 1.87 Teacher cleans, decorates room . 0.18 0.54 Teacher writes on, decorates board 0.67 1.17 Teacher leaves, enters room 0.79 1.20 1.43 1.74 Teacher talks with visitor aBased on eighteen observations of 178 teachers. 232 Table A.3.——Means and Standard Deviations of OScAR Cate- gories: Grouping Activitiesa Category Mean S.D. Selection of Members Teacher's choice of volunteers 0.83 1.59 Teacher's choice on some arbitrary basis 5.01 4.24 Pupil's choice of volunteers 0.17 0.50 Pupil's choice on some arbitrary basis 0.37 0.86 Pupil goes to group of his choice 0.42 1.18 Individual task with teacher Over 1/2 the class with teacher 3.19 3.81 Four to 1/2 the class with teacher 0.83 1.80 Two to three with teacher 0.77 1.43 Group task with teacher Over 1/2 the class with teacher 7.32 4.18 Four to 1/2 the class with teacher 3.15 3.13 Two to three with teacher 0.55 0.96 Individual task without teacher Over 1/2 the class without teacher 4.76 3.65 Four to 1/2 the class without teacher 0.94 1.58 Two to three without teacher 0.62 1.20 Group task without teacher Over 1/2 the class without teacher 1.39 2.42 Four to 1/2 the class without teacher 0.54 1.06 0.58 1.34 Two to three without teacher aBased on eighteen observati ons for each of 163 teachers. 233 Table A.4.——Means and Standard Deviations of OScAR Vari- ables: Materials Used by the Teacher and Pupilsa Teacher Pupils Mean S.D. Mean S.D. Blackboard 4.55 3.00 1.94 2.63 Map, chart, picture 1.04 1.57 1.18 1.89 Slide, film, etc. 0.76 1.39 0.28 0.93 Audio aid 0.44 1.05 0.38 1.28 Three dimensional object 0.89 1.64 1.51 2.15 Text, workbook 6.95 4.01 8.48 4.30 Supplementary reading material 1.44 1.79 3.31 3.06 Writing 1.33 2.34 7.14 4.77 Handicraft, art 0.65 1.38 2.13 2.40 Special teaching aid 1.88 2.34 Special learning aid 2.64 2.98 No materials used 2.97 2.69 1.86 2.13 aBased on eighteen observations in each of 178 classrooms. Table A.5.——Means and Standard Deviations of OScAR Vari— ables: Subject Matter Observeda Subject Matter Mean S.D. Reading 4.65 3.66 Mathematics 3.76 2.89 Language Arts 4.91 3.33 Social Studies 2.22 2.76 Science 1.72 2.59 Health, recreation 0.87 1.67 Arts and crafts 1.17 1.96 Music 0.33 0.86 Test 0.59 1.10 Other 0.57 1.65 aBased on eighteen observations for each of 178 teachers. 234 Table A.6.——Means and Standard Deviations; Personality Variables. Variable Mean Standard Deviation Achievement 12.65 3.89 Deference 12.42 3.81 Order 10.74 4.32 Exhibition 13.85 3.54 Autonomy 12.29 4.49 Affiliation 16.47 4.27 Intraception 17.98 4.89 Succorance 12.35 4.69 Dominance 13.36 4.52 Abasement 15.71 4.77 Nurturance 16.44 5.00 Change 17.24 4.71 Endurance 13.17 .5.08 Heterosexuality 13.16 5.81 Aggression 11.11 4.46 Consistency 11.31 1.96 MTAI 39.88 25.28 63.21 11.92 Dogmatism APPENDIX E FACTOR LOADINGS FOR ONE HUNDRED CLASSROOM OBSERVATION VARIABLES ON EACH OF FIVE FACTORS IN THE FOURTH ROTATED SOLUTION val—raw...» 1,, . 236 Table A.7.——Factor Loadings for One Hundred Classroom Observation Variables on Each of Five Factors in the Fourth Rotated Solution. Factor Loading Variable h2 Number I II III IV v 1 346* 607* 137 189 147 564 2 136 089 540* 453* 063 527 3 054 100 001 -069 442* 213 4 —012 ,111 -184 —025 489* 286 5 —236 —054 —174 —193 399* 285 6 —216 025 112 —078 485* 302 7 506* 231- 148 175 —120 376 8 479* —011 —084 015 —055 239 9 325* —173 561* 072 —034 456 10 —l84 009 025 005 249 097 ‘ 11 131 198 011 —078 114 076 12 393* 227 236-' 062 —040 267 13 —032 634* —044 -093 033 414 14 407* 226 -015 101 —053 230 15 694* -051 —018 135 —186 538 16 —037 —055 —215 431* 083 244 17 237 018 -047 —059 106 074 18 437* —033 —222 029 —238 299 19 -197 —108 258* 103 209 172 20 105 -207 056 077 060 067 21 —028 —024 —121 —l60 077 -048 22 —108 —246 166 056 404* 266 23 190 —256 131 -029 206 162 24 -121 -165 —091 —460* 376* 404 25 —035 —151 -197 —358* 255 -256 *Significant at the .01 level. 237 Table A.7.—-Cont. Factor Loading \ Variable 2 Number h I II III IV V 26 114 145 —065 —118 667* 498 27 —120 —132 —115 —065 487* 287 28 ~238 —041 105 -032 557* 380 29 576* —154 044 163 056 388 30 172 —052 105 242 —100 112 31 202 348* 023 101 -228 225 32 ~168 —168 070 255 —050 129 33 —092 621* 115 —160 157 457 34 357* 111 055 466* —203 401 35 065 —219 —040 —165 -282* 161 36 323* —341* ~075 -044 -276* 304 37 450* 082 —051 —025 —166 240 38 138 —192 072 208 —312* 202 39 213 -l95 —170 —020 —271* 186 40 241 095 —023 101 —045 080 41 —148 121 613* —038 087 421 42 —199 234 591* —033 154 469} 43 459* —072 363* —303* —142 460 44 136 381* 062 —033 -114 181 45 169 ~235 131 062 —131 122 46 283* —252 201 —044 122 201 47 267* -045 —018 -007 303* 165 48 203 —074 232 -007 240 159 49 088 —291* —023 —133 448* 311 50 173 —109 083 —116 577* 395 51 553* —244 -007 135 -008 383 52 057 ~129 —217 —334* -115 192 53 —040 024 101 —314* 075 116 54 290* 045 052 017 063 093 55 075 —122 073 —162 —319* 154 _______._——————- *Significant at the .01 level. _ 'W‘vajvw‘civx‘ 238 Table A.7.-—Cont. Factor Loading Variable 2 Number h I II III IV V 56 188 —224 063 198 —104 139 57 —047 —094 247 049 062 078 58 —003 153 034 —246 050 087 59 252 —l69 052 —080 —092 109 60 099 089 —568* 062 017 344 61 —018 341* 223 —387* —000 316 62 023 —020 051 081 —102 020 63 101 —008 467* —227 012 280 64 064 —031 336* —174 —095 157 65 —005 070 024 —263 124 090 66 —049 —021 —025 —321* —309* 202 67 120 —123 217 —039 054 081 68 —122 —084 094 —053 111 046 I 69 —032 613* -134 —186 051 433 70 524* 147 008 108 —130 325 71 —062 ~077 —224 460* 135 290 72 208 —036 —O35 050 —038 050 73 579* 080 -053 154 038 370 74 *448* 611* 025 —120 109 600 75 271* 103 160 —071 196 153 76 257 300* 226 —086 —047 216 77 488* —166 —043 —186 068 307 78 174 ~038 —029 374* —069 177 79 —091 —612* 178 —079 -147 442 80 147 .490* -008 -185 —086 303 81 199 —041 365* 136 —210 237 82 —042 —061 ~082 347* 189 168 83 062 -126 235 —024 —182 109 84 536* —145 —013 122 —139 343 85 —506* 412* 332* —125 087 559 _________.—————— *Significant at the .01 level. Table A.7.——Cont. 239 Factor Loading Variable 2 Number h II III IV V 86 -036 ~068 641* —120 029 432 87 —064 273* 584* —093 006 429 88 445* —l85 266* —340* —l70 447 89 192 —131 187 489* —O99 338 90 157 —286* —3l7* —071 112 225 91 —043 319* 266* —497* 057 425 92 -l39 284* 218 048 -133 168 93 105 —061 037 -l32 034 035 94 —128 —151 243 201 —024 139 95 045 —073 —043 503* —019 262 96 246 —238 —385* —039* —093 275 97 587* —217 —005 —234 004‘ 447 98 232 —189 —310* 030 —189 223 99 ~263* —O9l 132 155 020 119 100 —008 022 104 037 086 020 *Significant at the .01 level. The Guilford—Lacy expression was applied to compute the standard error of factor loadings.1 This is justifiable since the factor loading is a function of the standard error of the original correlation matrix. 1J. P. Guilford and J. I. Lacy, Printed Classi- fication Tests, Army Air Forces Aviation Psychological Program, Research Report 5 (Washington, D.C.: Government Printing Office, 1947), 919. 240 First, the Standard Error (S.E.) is determined by taking the reciprocal of the square root of the number of variables analyzed. 1 .E. = » S m“ 1 S.E. — 100 = .10 Using a fiducial level of .01, significant factor loadings equal or exceed (.10) X (2.58) or 0.258. Those factor loadings marked with an asterisk meet the above criterion. Twenty—feur of the one hundred variables were not significant on any factor. These variables might well be eliminated in further research. On the other hand, seventeen variables correlated Significantly with more than one factor.